This item is only available as the following downloads:
The Cell Phone: Calling and Texting Behaviors between Close Relationship Partners By A Thesis Submitted to the Division of So cial Sciences New College of Florida In partial fulfillment of the requirements for the degree Bachelor of Arts with a focus in Psychology Under the sponsorship of Dr. Michelle Barton Sarasota, Florida April, 2013
i Acknowledgements I would like to profusely thank Dr. Michelle Barton, with whom I could not have done this project. She helped me along every step of the way and never gave up on me even when I wanted to give up on myself. Her guidance and composure gave me the courage to keep moving forward. I would like thank Prof. Duff Cooper, the Statistics Superman, who stood by my side and patiently assisted with the painstaking i nferential statistical analyses. Some days his silly jokes and excitement for math were what gave me the en ergy to press on when I wanted to throw in the towel. I would like to thank m y mom, who not only gave me the food and resources to carry on in my thesis process, but also provided me with all the warmest over the phone hugs and kisses, letting me know ever ything would be alright even when I could not even picture being able to complete this enormous task. I would like to thank m y sister, who gave me tremendous support throughout this whole process and beneficial edits at the end. W hether located in Orlando or France or elsewhere, my compassionate sister was always willing to listen to me grip e about my thesis tribulations and yet somehow always managed to have me leave our conversations full of hope and a sense of peace. Finally, I would like to thank m y bes t friend Diana Watson for always allowing me to interrupt whatever she was doing to have a ch at. Sometimes it was to verbal ly work out a thesis problem, other times it was merely to have a light hearted conversation for the sake of not thinking about thesi s. Your constant provided me with the power to push through and get it done
ii TABLE OF CONTENTS ACKNOWLEDGEMENTS i TABLE OF CONTENTS ii LIST OF TABLES iv LIST OF FIGURES v ABSTRACT vi INTRODUCTION 1 Computer Mediated Communication 3 The Cell Phone: Calling and Texting 5 Reasons for Communication 9 Demographics: Gender and Age 14 Close Relationships 19 The Current Study 24 METHOD 29 Participants 29 Materials 29 Measures 29 Procedure 34 RESULTS 36 Descriptive Results of Coded Reasons for Communication Choice 36 Actual vs Perceived Frequencies 37 Age Correlations 38 Distance Correlations 38 Gender Frequencies: Two way ANOVA 39
iii Mode Frequencies: Three way ANOVA 40 DISCUSSION 42 Partner Type 43 Mode 44 Interactions 45 Age 46 Gende r 47 Physical Distance 47 Co mmunication Mode Reasons 48 Actual vs. Perceived Communication 48 Limitations 50 Future Directions 5 1 Conclusion 51 REFERENCES 5 3 TABLES 57 FIGURES 6 2 APPENDIX 76
iv List of Tables TABLE 1: Reason Code Frequency Distribution by Partner and Mode TABLE 2: Pearson Correlations of Partner by Mode for Actual vs Perceived Frequencies TABLE 3: Pearson Correlations of Actual Calling/Texting Frequencies by Partner's Age TABLE 4: Pearson Correlations of Actual Calling/Texting Frequencies by Partner's Physical Distance
v List of Figures FIGURE 1: Graph of Calling and Texting Frequencies by Gender (with outliers) FIGURE 2: Graph of Calling and Texting Frequencies by Gender (without outliers) FIGURE 3: Graph of Communication Frequencies by Partner Type FIGURE 4: Graph of Communication Initiation Frequencies across Partner Type FIGURE 5: Graph of Communication Initiation Frequencies by Mode FIGURE 6: Grap h of Mode Frequencies by Partner Type FIGURE 7: Graph of Calling Frequencies by Initiator across Partner Type FIGURE 8: Graph of Texting Frequencies by Initiator across Partner Type FIGURE 9 : Graph of Communication Frequencies by Partner Type (without Siblings) FIGURE 10 : Graph of Communication Initiation Frequencies across Partner Type (without Siblings) FIGURE 11 : Graph of Communication Initiation Frequencies by Mode (without siblings) FIGURE 12 : Graph of Mode Frequencies by Partner Type (wi thout Siblings) FIGURE 13: Graph of Calling Frequencies by Initiator across Partner Type (without Siblings) FIGURE 14: Graph of Texting Frequencies by Initiator across Partner Type (without Siblings)
vi THE CELL PHONE: CALLING AND TEXTING BEHAVIORS BETWEEN CLOSE RELATIONSHIP PARTNERS Stephanie Kownacki New College of Florida, 2013 ABSTRACT The calling and texting behaviors of college students were studied using information s age and sex demographics, and reasons for communication mode choice. Additionally, an examination of the validity of self reported communication data was conducted via comparisons with actual calling and texting logs. Students completed a survey asking about their perceived commun ication with four close relationship partners (parent, sibling, intimate partner, and best friend). Also, the survey collected data on the age and sex of both interlocutors, the physical distance between them, and reasons why the participant would choose t o call vs text, or text vs call. Then, actual calling and texting logs from the participant's cell phone were recorded. Main effects of partner type and mode type indicated intimate partners to have the highest frequency of communication and texting to be the primary mode of communication. Age of parents was positively correlated with the frequency of calling. The validity of self report data was inconclusive. These findings suggest that new advances in technology are changing the way we choose to communica te and those choices are possibly influenced by whom a person wants to contact __________________________ Dr. Michelle Barton Division of Social Sciences
1 The Cell Phone: Calling and Texting Capabilities as seen between Close Relationships only have we given up on writing letters to each other, we barely even talk to each other. People have become so accustomed to texting th Seriously...I'm Kidding (p.138 140) This quote by comedian and talk show host Ellen DeGeneres presents a clea r picture of how much cell phones have become integrated into our everyday lives, changing the way we communicate, which has in turn aided in the slow demise of some age old methods of communication. Text messaging is hyperbolically highlighted as the up a nd coming norm for basic communication, although calling is still seen as the primary mode of communication at the present (Ishii, 2006; Auter, 2007). Cell phones started to become a mainstay around the beginning of the 21st century, with records showing cell phone ownership among teens increased from 10% to just under half at 46% between 2002 and 2005 (Lenhart, Madden, &Hitlin, 2005). Since the emergence of the cell phone there has been a steady decrease in the use of landlines (Ishii, 2006). Cell phone use has been seen everywhere from school and work, to church and on the bus (Raacke& Bonds Raacke, 2005).
2 Mobile phones have become an integral way to organize life because of all the conveniences they allow. Besides the advantage of mobility, with a cell phone it is possible to schedule appointments while on the go or to give information to parents and friends at the same time; these phones connect us in ways that were never possible before (Kim, Kim, Park & Rice, 2007; Lenhart, Ling, Campbell, & Purcell, 2010). Cell phones are used to communicate with those who are important in our lives to talk about serious issues, as well as to communicate with schoolmates/friends about less pressing issues (Ling &Stald, 2010). For example, in a focus group discussion, one teenage girl remarked Text messaging has seen positive reception from many users especially because of the covert nature the message can afford. Many teens find thi s aspect to be important for times when they do not know if the other person is busy, such as in a meeting, when a call would be inappropriate (Raacke& Bonds Raacke, 2005; Lenhart, Ling, Campbell, & Purcell, 2010). However, texting can be tedious if there is a long message to be conveyed, and so, for many instances calling is still seen to be preferable to sending a text, resulting in a trend of fewer calls occurring, but more time spent talking (Lico pp e&Smoreda, 2005). Characteristics of the person who is performing the communication have been shown to possibly affect communication behaviors. Females have been shown to communicate more than males in general, with teen girls sometimes displaying double the rates (or more) of daily texts than males (Pierce, 2 009; Lenhart, Ling, Campbell, & Purcell, 2010). Also, mothers were found to call their children more than fathers (Kennedy & Wellman, 2007). In 2012, s urvey data taken from university students, ages
3 17 to 35, revealed an inverse correlation of age with the frequency of text messaging (Jin & Park ). Although, cell phones are used for communication needs in general, the majority of its use occurs in close relationships such as families, romantic couples, and friends. These intimate communications over cell pho nes are even thought to enhance the relationships themselves (Jin & Park, 2010; Ishii, 2006; Kim, Kim, Park, & Rice, 2007). With the dramatic impact the cell phone has made changing where and how and why we communicate, new understanding of what drives co mmunication behaviors is needed. Some literature has examined communication habits by partner type, by mode, by gender, and by initiator. Yet, it is still unknown how mode choice is actually affected by all of these variables because they need to be explor ed together to gain a comprehensive picture of the influences behind communication choices. The following literature review will provide a basic overview of the emerging trends guiding how the cell phone is used for calling vs. texting. An initial look at how digital communication has introduced mobile, and consequently some text based, communication into the everyday lives of people in the 21st century is presented briefly Then, a focused look at the calling and texting norms that have surfaced in the mos t recent years is explored. After that, reasons why someone would choose to call or text is examined. Following that, cell phone use across age and sex demographics is reported. Finally, some literature focusing on calling and texting behaviors observed in close relationships is reviewed. Computer Mediated Communication Communication can take many forms. In the past, there was a heavy reliance on face to face communication and letter writing, but in modern times, digital technology
4 has drastically expanded the variety of ways an individual can communicate with someone else. As well as expanding the ways people connect, these new technologies have allowed an expansion of the distance between two people and integrated text based modes in daily life that bring their own pros and cons into the decision making process of choosing a communication mode. Computer mediated communication (CMC), also sometimes are referred to as Socially Interactive Technologies (SIT) (Pierce, 2009) or Information and Communication Tech nologies (ICT) (Licoppe & Smoreda, 2005), all include essentially electronic, often symbolic, text based, technologies involving people engaged in a process of message interchange (Perry & Werner Wilson, 2011). The major forms of communication that fall unde r this umbrella of CMC are found within cell phones and computers, and include social networking sites, text messaging, instant messaging (IM), and e mail (Pierce, 2009). In an analysis of three empirical studies, Licoppe and Smoreda (2005) scrutinized how social networks are composed, the ways that communicators choose to create a message, and the available communication means. The overall conclusion that arose from their research was that these communication technologies Just as CMCs each have a unique composition of characteristics, they are also distinguished by their degree of synchronicity. Each communication mode can vary in the spectrum of synchronicity, from synchronous to asynchronous. Synchronous communication happens in real time with no extended gaps in the flow of conversation, McKenna and Bargh(2000), as cited by Mandell and Muncer (2007). Some well known
5 synchronous communication exchanges include phone calls and instant messaging when responses are immediate. On the other hand, asynchronous communication has gaps in the flow of conversation where minutes, hours, or days can pass be fore a response is provided. Popular examples of asynchronous modes of communication include texting, e mailing, and traditional letter writing. Often, synchronous communication is used when communicators need the response information quickly or when a wea lth of information needs to be relayed. Asynchronous communication media are chosen when the person being communicated with is busy or the person making the communication wants to allow time to plan out their response so they can fully articulate their mes sage (Mandell&Muncer, 2007). Specifically, it has been mentioned that teenagers like the asynchronous interactions of texting because it can be used casually to convey messages. Also, since texting is not in the moment, it was suggested that text messaging can put the texter at ease when communicating with potential romantic partners (Lenhart et al., 2010). As well, convenience and lack of intrusiveness have been associated with why asynchronous communication is seen as useful (Kennedy & Wellman, 2007). An xiety from low social skills can sometimes be provoked by synchronous communication, like talking on the phone, and so people also sometimes choose to use asynchronous communication modes to avoid personal weaknesses (Ishii, 2006). The Cell Phone: Calling and Texting An overview of usage patterns by frequency of calling and texting will initiate the analyses that are to follow. Both calling and texting occur at high rates, and research shows differing rates of use. For example Ling and Stald (2010) conducte d a large scale
6 survey of how mobile communications impac t social dynamics with intimate 1 partners. Of the 2,772 people surveyed, 80% reported that they chose to use voice calls and 79% of the sampled population also chose to use texting when communicating with close people in their lives on a daily basis. Since there is an expanding array of features and applications included in modern cell phones, Auter (2007) wanted to gauge how university students allot their time when they use their cell phones. A surv ey was used to collect information about how time is spent on their cell phone, categorizing time in three ways: calling and voice mail, extra services that deduct from monthly allotted minutes (e.g. downloading files, texting, internet access), and applic ations that do not use monthly minutes. From this data collection it was shown that even given the wide variety of alternative ways to use their cell phone, approximately 90% of the time on the phone was occupied with making calls. Alternatively, calling w as not a common mode chosen in the study performed by Kamibeppu and Sugiura (2005), looking at how mobile phones were used and perceived by Japanese 8th graders. The researchers gave questionnaires to teachers to pass out in class. When all the data were c ollected, the study showed that 54.3% of the students who owned cell phones sent or received over 10 texts a day to their friends in close physical within a week. Mob ile communications use was the focus of a Japanese nationwide two wave panel survey conducted by Ishii (2006). The survey collected information in late 2001 and 2003 pertaining to how much Japanese citizens used landlines, the calling function 1 It was noted that the term intimate was intended to mean close relationships, not sexual ones.
7 on cell phon es, text messaging, and sending e mails on a computer, along with demographics for those surveyed. From the longitudinal information received, Ishii observed that the percentage of cell phone owners went from 62.9% in 2001 to 74.5% in 2003. Along those sam e lines, text message users were also found to rise from 54.5% in 2001 to 72.6% in 2003, showing the explosive growth of cell phones and more importantly of texting that has happened in the recent decade. Another, more recent comparison of texting across r ecent years was conducted by Lenhart, Ling, Campbell, and Purcell (2010) as part of ongoing research for the Pew Department of Communication Studies. To stay current on national trends in America, phone surveys and focus groups gathered data about cell phone use (primarily texting) of a randomly selected 800 teenagers aged 12 17 and one of their parents. The frequency of calls among teens was found to be about five calls made or received per day. From their research, daily texting was a common occurrence among teens. In fact, they saw a 16% increase in the proportion of teens who text message every day, rising from 38% in the spring of 2008 to 54% in the fall of 2009. The addition to these results, evidence a stable trend in texting occurring globally. Students are often quite familiar with, and frequently use, cell phones for communication purposes. Ishii (2006) collected data on the oc cupation of research participants. Analysis of the occupation data revealed that there was a negative correlation between cell phones for calling and being a student. However, this student status was positively correlated with the use of texting. Given thi s finding, Ishii
8 Having college students as a target group, Skierkowski and Wood (2012) focused th is age demographic is prone to using text messaging as their primary mode of communication. College students were recruited to fill out an online survey three times a day for a 5 day period to gather information on their daily usage of texting, Facebook, e mail, calling instant messaging and face to face interactions. The survey also asked who was being communicated with via each communication mode. When analyses were run, their hypothesis was supported in finding that the college aged sample used texting most often as their communication mode. This research along with the work of Ishii (2006) suggests of a consistent trend between being a student and mostly choosing to use text messagi ng. This could be a result of students being in class much of the time so they mostly rely on texting, but more research is needed before any conclusions can be made. chosen from a voter list gathered da ta to compare texting use and e mailing use The rese archers measured use of texting use of e mailing, social network diversity, and social were run, overall findings i ndicated that, unlike computer e mail, text messaging was providing direct contact. In conjunction with this finding, from their survey study of 8th graders use and feelings about cell phone s Kamibeppu and Sugiura (2005) similarly proposed that text messaging on cell phones builds stronger friendships. From participant reports, the authors actually suggested that text messaging may even enhance face to face relationships.
9 Rea sons for Communication In trying to fully understand the uses and tendencies of calling and texting there comprehensive understanding of the dynamics at work, some research ers have instead chosen to bypass measuring communication behaviors and research the reasons and motivations for the behavioral choice between calling and texting instead. Some researchers have specifically studied text messaging. For example, in an explor ation of text messaging use, Shadyad, Pakdaman, Hiedary, HasanZadeh, MirnaderMiri, Asadi, and ShirAlipour (2011) collectedresponses from a questionnaire studying texting motivations, frequency, and con tent. They conducted a factor analysis which teased apa rt the overarching reasons for why people use this mode of communication. The analysis revealed four core themes. The first entailed motivation s for activities that are socially acceptable, for example using texting to contact a loved one. being, suggesting that this makes the recipi ent feel more secure. The fourth theme, to skills. Although limited, some comparisons to calling reasons can be made about concern for money. Also, the direct mention of fa ce to face communication being avoided provides more indirect knowledge of alternative choices. In another texting focused study, Raacke and Bonds Raacke (2011) used questionnaires to examine text messaging, specifically targeting the categories that
10 motiv ate use. Three main dimensions emerged for texting: communication to share, communication to avoid, and covert communication. The first dimension, communication to share, was composed of sharing information, getting information, communicating with multiple people at one time, keeping in touch with friends, ease of use, and speed of communication. The second dimension, communication to avoid, included when communication was used to avoid having a verbal conversation or avoid having a confrontation or an emot ional situation. The third and last dimension, covert communication, was about communication when participants could not make calls at the time or the other person could not talk at the time of contact. The findings gathered from these two studies highligh t some motivations to consider researching more. Their results are valuable to a study about communication reasons, but they must be used and interpreted cautiously because the content only provides indirect avenues for comparison of texting motivations ag ainst calling motivations. These findings can act as springboards for other research, particularly the need to directly address alternatives. For results and analyses have been curtaile d, leaving out information about texting exposing only fractions of the relationships between variables. Included in their investigation of communication technologies were phone calls, text messaging, e mailing, and use of social networking sites (SNS), bu t the authors only provided results and discussion on phone calls and SNS. Gentzler et al. commented on the notion that levels of feelings of communication mode frequencies for interactions with their parents. The data revealed numerous positive relationship quality indicators related to
11 calling, such as intimacy, support, greater satisfaction, and instrumental aid, when communicating with pare nts. This led Gentzler et al. to suggest that phone calls may create a feeling of connected with their parents, yet still be independent. Given these results, phone calls seem to be link ed to creating a dynamic that allows for a preferable amount of intimacy. S ome research has delved into the field of motivations behind why a n interlocutor chooses a mode of communication based on analysis of various modes of communication. At times, when presented with a choice a person is subconsciously driven by considerations unique to that person, and other times one might consciously contemplate a choice based on the underlying purpose of the communicated content. Sometimes, part of the decision makin g process of communication mode is influenced by the personal, and sometimes psychological, traits of the communicator. Everyone is shy or uncomfortable in social situations at some point in their lives. For a portion of the population, low social skills o r general anxiety affect how these people interact with others. For example, Hashimoto (2002, as cited by Ishii 2006) conducted research of Japanese youth that showed many teens had anxiety about direct communication. Because anxiety deterred some from cal ling, text messaging was found to be the alternative choice for communicating in this study. This finding led Ishii to conduct a study on mobile communications, with particular attention paid to examining r to differences in communication mode habits. In her own study, Ishii hypothesized that those participants deemed as having low social skills would be more inclined to use text messaging over phone calling as a way to
12 get past social incompetencies. A two wave, nation wide survey was taken of Japanese citizens. The communication modes studied were landlines, cell phones for calling, cell phones for texting, and e mailing. When Ishii analyzed the data, findings indicated tha t the level of participants l skills was a significant predictor of the amount of calling participants made. Hence, when a participant was found to have high social skills,that participant was also found to have high rates of calling. The level of social skills was not initially related to text message usage, but when age and gender we re controlled, the study found a negative correlation between social skills and the inclination of a participant to use text messaging. More recently, Pierce (2009) had studied the impact social anxiety can play in the communication behaviors of teens using SITs. Pierce hypothesized a correlation between feeling un comfortable communicating in person and feeling more comfortable texting. This was based on the established research indicating that girls are more prone to having social anxiety, and that girls are more at ease during online interactions rather than face to face ones (Hamburger and Artzi, 2000, as cited by Pierce, 2009). A study was conducted using 280 high sch ool students, boys and girls, aged 14 20. The questionnaire in the study collected data on demographics, general usage of SITs, and general comfort level using various communication media during interactions with others. The results of the study indicated that social anxiety during face to face interactions predicted more comfort communicating by texting. Because communication is an inherently social activity, a person who experiences social awkwardness might logically be influenced by this in their choic e of media when communicating. Not all people, however, find themselves taking the interpersonal aspect
13 into such high consideration, and instead of their mode selection being based on qualities of the self, it is based on the message that needs to be conv eyed. What needs to be said can play an influential role in the choice of communication. For instance, someone could analyze mode choices from a time perspective, looking at mode by speed if the message is time sensitive, as opposed to comparing modes by t heir allowance for time to plan out responses. Such mode capability perspective taking was witnessed by the participants in the following study. In 2007, Mandell and Muncer wanted to explore the reasons why adolescents use different communication media, su ch as the internet and cell phones, in their social lives. Using small focus groups of students 18 20 year olds, data were recorded and then analyzed using word analysis software to pull out the major themes that arose in the communications. The analysis i lluminated that the core motivation driving communication mode choice was the level of control each afforded. Text messaging was observed as being the preferred method of communication when participants felt like they wanted time to think about how they wa nted to respond and plan out their words, especially for emotional situations. The participants also reported using text messaging for limiting interactions when they were communicating with people they did not know well, as well as when they did not have much to say. The study found that using the phone to make calls wa s frequently initiated with the intention to have a long lives. Calling was also chosen when there was a n urgent matter to address. Differences between the modes of calling and texting are clearly demonstrated as being a common way to choose which mode to use.
14 Moreover, in 2005, Kamibeppu and Sugiura surveyed Japanese teenagers about their cell phone habits, including ownership patterns, feelings on the effects of their cell phones, and frequencie s of use of calling and texting 2 From the survey, calling was shown to be utilized often when there was a need to give important information. The the ease of sending a short message afforded by text messaging. Choosing to call is highly mot ivated by wanting to transmit lots of information, while texting is primarily motivated by convenience. Demographics: Sex and Age Sex and age are often important variables to look at in psychological research. Knowing the demographics about a person can give a researcher background into that ontext they may also be useful in shedding light on some of the communication choices people make. Kamibeppu and Sugiura (2005) investigated how often cell phones are used for texting purposes, how many are owned by junior high students, and how cell phone s in Japan are perceived as impacting relationships between friends, communication habits, and psychological well being. The researchers had questionnaires handed out to 8th graders in five public schools in Tokyo. They found a significant gender differenc e, in rates of ownership, with more girls owning cell phones than boys. More ownership lends to higher potential for calling and texting to occur at all and suggests more interest in communicating with others 2 (2006) that in Japan texting and e and so both are referred to distinguish between these two serv
15 nd how they are used by family members. They created a 32 page survey and conducted home interviews from a subsample of their whole sample of Canadians. Three hundred and fifty participants took the survey answering questions about their internet use, comm unity involvement, household relations, personal community networks, and social attitudes. The interview was used to gain more detailed information. They categorized participants by their ICT e apt to be light or non users of the internet compared to males. As well as this tendency in internet usage for women, mothers chose to use cell phones to call their children and romantic partners more often than fathers did. These results suggest that o lder females may try to avoid text based interactions and when they do communicate by calling, they report higher frequencies of communication than fathers. Raacke and Bonds Raacke (2011) investigated college frequency of Short Messa ging Services (SMS), also known as texting. The main goals of this study were to assess the prevalence of SMS use, where it is used and why it is used. One hundred students were recruited to complete a questionnaire asking about what type of phone they ow ned, the frequency of use, where SMS is used (e.g. work, home, driving, etc.), why they use SMS, and finally demographic questions. Between genders females were shown to use SMS more often while partaking in a variety of activities. Additionally, females were shown to send and to forward texts more often than males. Once again the results show more evidence that females tend to communicate more overall. W ith younger fem ales, text based communication wa s accepted and used, which
16 is a complementary finding with the results of the Kennedy and Wellman study, where the older females did not want to use text based communication. Pierce (2009) examined SITs, social networking sites (SNS), calling and texting on cell phones, and instant messaging (IM), in relation to social anxiety and how social students were recruited to complete a questionnaire, which included a demographics section, an overall SIT usage section and a sectio n covering comfort levels with various modes of communication (face to face (FTF), online, text messaging, and calling). Results indicated differences, including findings which again indicated that females used cell phones for calling and texting more than males. Also, females registered more social anxiety compared to males and were more comfortable talking online and via texting low social skills were associated with less willingness to make calls. As mentioned earlier, the Lenhart et al. (2010) study used survey data accompanied by focus group data to conduct research on the trends in teen cell phone use. Their review found that boys had a lower frequency of texting, sending and receiving texts, with approximately 30 texts a day as opposed to the 80 texts that girls in this study sent and received. When all the results were considered, the researchers concluded that there is a higher likelihood of girls to use both ca lling and texting than boys. Underwood, Rosen, More, Ehrenreich, and Gentsch (2012) examined adolescent digital communication mode habits, including e mailing and instant messaging, with a special focus on texting. The researchers took data gathered from a longitudinal s tudy in which Blackberries had been distributed to the participants to use. The experimenters
17 looked at a 2 day sample of text messages from their population of 175 15 year old adolescents, obtaining both billing records and questionnaire da ta. Surprisingly, when the researchers compared billing records and questionnaire data, they did not observe any documented sending and receiving a similar numbers of texts. Calling usage and outgoing text usage were moderately correlated positively for both genders. These findings of equal gender usage are in opposition to the other literature reported in the present examination, such as Lenhart et al. (2010), and Pier ce (2009), who found females doing more communicating than males. Additionally, these researchers also had the actual records of the communication behaviors, which spark interest in knowing if survey s and questionnaire sare accurate For an understanding of how calling and texting works in social networks to formulate loose or tight groups, Ling and Stald (2010) used questionnaire sampling of con tacts chosen knew each othe r the n the group was considered tight, while fewer mutual contacts were considered loose or diff use. Sex was found to inversely covary with group tightness indicating that males tended to have looser interactions than females. A more complex analysis found that age, face to face interactions, and cell phone calls positively interacted with tightness where the higher frequencies of calls and face to face interactions were shown to have the tightest groups, and this was shown to increase with age. With cell ph one use on the rise, the researchers Jin and Park (2010) were interested in what motivates people to use calling and texting on cell phones. To collect
18 information, an online survey was distributed to university students, ages 17 to 35, asking about their cell phone usage, interpersonal motives for cell phone use, face to face interactions, and loneliness. Both age and sex variables were found to have significant cor related to age, where the older someone was, the less they cited this motive. Another inverse correlation with age was found for the frequency of text messaging where higher texting rates were seen in younger participants and this declined as age increased In terms of gender, males were observed to have lower levels of both inclusion motives (e.g. eir female counterparts. In a large scale study by Ishii (2006), questionnaires were filled out by Japanese citizens, ages 12 69, to obtain data about their everyday communication media tendencies. The media referenced were landlines, calling on cell phone s, text messages, and e mails. From the 1,245 participants sampled, results revealed that males made cell phone calls at double the rates of the females in the sample. For females, texting was the most regularly used mode of communication, a finding especi ally true for teenaged girls paralleling the findings of Raacke and Bonds Raacke (2011). There was a positive correlation between age and amount of calling, where, as age increased so d id the frequency of calling. To see if calling and texting frequencie s varied with sex and age, Ishii ran an ANOVA. Individual main effects of age and gender were seen in the positive direc tion for cell phone calls. I ndividual and interaction effects arose for text messaging, highlighting the strength of the association bet ween young females and their use of
19 discovered some absolute age difference s. If dyad communicators were within the same age bracket they tended to use text messaging, but when the absolute differences in age were in different age brackets, landlines were indicated as being used. Summary. Overall, these studies give a clear indication that the variables of gender and age are predictors of communication behaviors, with the exception of whole, these studies have shown that females tend to do more communication behaviors from higher ownership to higher rates of calling and texting ( Kamibeppu & Sugiura 2005; Pierce, 2009 ;Lenhart et al., 2010 ). Males seem to be willing to call more than females, but from Jin an d Park (2010 ) results, there seems to be a general preference toward calling and a decrease in the preference to text as age increases. Over both age and sex, Raacke and Bonds Raacke (2011 ) data indicate that the lower age of fe males is shown to be associated with a preference toward texting. On the individual level, age and gender are important aspects to keep in mind in communication research. However, another important variable of research may come from looking at the communic ating dyad and who they are in relation to each other. Close Relationships Cell phone communication patterns can be susceptible to the influence of variables like the gender or age of the people communicating. However, another important factor to take into consideration is the dynamic of the relationship of the
20 interlocutors In 2010 Ling and Stald addressed the question of how m obile communications and ICTs influence social relations and social cohesion, especially within intimate spheres. They conducted a survey asking Norwegians to name up to five people with the participant's relationships with the contacts, how many of the contacts also knew the other contacts, which modes of communication were used, and how often the modes were used. Tightness of a group was determined by how many of the individuals named al so knew each other. If all contacts knew each other, it was an extremely tight group, and if all the contacts were strangers, it was considered a diffuse group. Of the sample, 80% of those surveyed used call ing to talk to their intimate partners on a daily basis. The researchers found that mobile calling and face to with tightly knit tightness was not shown to be associated with high levels of text messaging. This suggests that calling and texting via mobile phones might be very influenti al in keeping groups together in general. As noted previously, Gentzler et al. (2010) conducted a survey of college ed to the beneficial qualities within the relationship: greater satisfaction, intimacy, support, and instrumental aid. These findings show that parents play an important role in calling behaviors. Although data were gathered on text messaging, no findings were presented on the frequencies or relationship
21 qualities associated with texting, leaving no comparison of the differences in relationship effects between those who call and those who text. Coyne, Stockdale, Busby, Iverson, and Grant (2011) conducted re search looking at romantic relationships and communication technologies. The researchers used a questionnaire to collect data about which media were used, the frequency of use of the media, and subsequent positive and negative communication by that medium. The never ) 5( very often ) likert scale. The negative communication measure rated statements on a 1( never ) 5( very ofte n ) likert scale. For frequency of media use, the researchers found that calls on a cell phone had the choices being communication via cell phone calls, texting, e mailing, s ocial networking sites, instant messaging, blogs and webcams. Also, out of all the types of communication positive and negative communication. The negative communication that was found had a positive nature, such as expressing affection. In all, cell phones were revealed to be the primary means of communication with romantic partn ers. Baym, Zhang, Kunkel, Ledbetter, and Lin (2007) examined relationship quality and media use in interpersonal relationships. Surveys of college students were used to look at interactions between four relationship types (acquaintances, friends, romantic partners, and family) and three modes of interactions (face to face, telephone, and internet). Each relationship type and mode of interaction was controlled, resulting in 12
22 versions of the survey which were randomly assigned to participants. The results s howed that telephone interactions with family members occurred in greater proportions than telephone interactions with friends, acquaintances, or romantic partners. Family members are some of the most important people in many lives and so it makes sense th at they would show to be important in communication habits as well. This study leaves out texting so that mode of communication remains to be tested across the four partner types. Only looking at calling and texting, Kamibeppu and Sugiura (2005) conducted a Japanese cell phone study of 8th grade students. The researchers collected information on which students owned phones, who they contacted and how, and their feelings about how their phones affected their lives. When the data were analyzed, the researcher s found that contacted primarily their families when they choose to use the calling function of their cell phone. However, there is no distinction given between parents vs. siblings vs. other family members when family is mentioned. Also, results showed that communication between friends was by and large conducted by text messaging. This preferred way to contact those particular partners. Moreover, this study shows that friends are another important type of communication partner and rank especially high when it comes to percentage of text messages sent and received. The research on the national cell p hone habits of teens by Lenhart et al. (2010) for the Pew Research Center provides more insight. According to their research, calling is a crucial method of communication between parents and teens. When the researchers studied the call records, they concluded that:
23 likely to be calling parents, with 68% of teens with cell phones saying they talk to their parents on their cell phone at least once a day. Talking with friends is a close second to parents, with 59% of teens with cell phones saying they talk with friends once a day or more often. About half of teens who have a boyfriend or girlfriend call them on a daily basis. Brothers, sisters and other family members are the least likely to be called on daily basis, with just about a third of teens who have siblings (33 %) saying The Pew results about texting revealed that out of the choice of parent, boy/girlfriend, sibling/other family member, or friends, teens answered that friends were texted the most, with 75% responding that they texted friends several times a day. Or, if a friend was the intended recipient, it was more probable that a text instead of a call would be chosen according to two thirds of their sample. Summary. Recognition that relationship dynamics have significant pot ential to literature, many studies show that calls are the primary mode of communication chosen when children communicate with parents. Another consistent finding among the research is that friends are the primary recipients of text messages. Minimal research looking at romantic partner communications leaves much room for further exploration to investigate common patterns in this area. What also remains largely unknown are t he typical communication behaviors between siblings. Though only a portion of all of the communication research available was reviewed the communication studies of Le nhart et al. (2010), and Baym et al. (2007), outlined above took into account a variety o f
24 relationships for comparison of communication behaviors, thereby creating a more comprehensive picture of main patterns of communication behaviors and choices in different relationships. The Current Study Existing communication studies have focused on co mmunication with a few close relationship partners, like parents and siblings (Goby, 2011), or on a singular intimate tie, such as a peer friendship (Kamibeppu&Sugiura, 2005). These types of studies add to the development of communication research, yet the y only capture sections of the major communication habits that use all four of the collectively implicated close relationship partners of college students: parent, sibling, int imate partnerand best friend While others have conducted research observing communication patterns by mode choice (Pierce, 2009), research is still needed to understand if communicating dyads (college students with the close relationship partners above) have similar or complementary behaviors across different communication media and who initiates contact in those relationships. Additionally, age and sex demographics are usually mentioned, but with the four main close relationships identified, more research needs to look at the demographics for patterns across each of these important partners. In the past, technology lim ited many researchers to rely on surveys and questionnaires when studying communication behaviors. T he information gathered pertained to com munication behaviors, however, questionnaires could only record mere perceptions of those studied, perceptions which were susceptible to the errors of human memory. Now in the digital age, a few researchers are attempting to collect actual logs of
25 data in stead of perceptions (e.g., Underwood et al., 2012). More studies and documentations of true records are required to give credence to the findings from earlier communication surveys, since we do not yet know how accurate self reports from older studies are Another issue to examine when studying communication choices is the communicate or wants instant clarification in a conversation, he or she might choose a synchronous commu nication medium to talk, where there are no delays in communication because it is in real time. Other times, people prefer to have some time to plan out what they want to say, so they might choose an asynchronous communication medium, where simultaneous ex changes do not occur, often where minutes, hours, or days go by before a reply occurs. Minimal research has been committed to looking into communication mode. This is especially true of synchronicity research focusing on calling and texting; therefore more research must be done to fill in these gaps. With all of these issues in mind, the present study sought to gather real calling/texting behaviors, as well as self repo rted perceptions, in pursuit of answering some of the questions that remain. The questions of interest in the current study included: How does the frequency of use of texting differ from frequency of calling as a function of the close relationship partner (parent, sibling, best friend, intimate partner)? Does the frequency of calling/texting differ as a function of the age/sex of partners? How accurate are perceptions of calling and texting compared to actual habits? Is the synchronicity of
26 the communicatio n a major determinant of whether someone will use one mode over the other? To address these questions, the current study used survey data and calling/texting logs to analyze the communication habits of college students with four close relation ship partners (parent, sibling, intimate partner best fri end). The participants answered survey questions that measured the perceived communication frequency of calling and of texting with each of the close relationship partners for a 1 week time period. For context, the survey collected the age, sex, and physical distance apart between each of the close relationship partners and the participant. Descriptive data were assessed in the survey by asking why the participant would text instead of call and call instead of te xt to each of the close relationship contacts. Then, the cell phone of the participant was used to record actual calling and texting logs from the previous week for each of the close relationship contacts. The log data included the time and date of each of the calls/texts, which partner was being contacted, and who initiated the communication. Hypotheses The first hypothesis expected that age of the partner would be positively correlated with call ing and negatively correlated with texting In support of this hypothesis, when Jin and Park (2010) surveyed college students ages 17 35 they witnessed negative correlations between the frequency of texting and age of participant. Another reported finding that is used as support for the hypothesis comes from research by Ishii (2006). From the nationwide Japanese survey, the age of communicator was also observed being inversely related to the use of the text mes sage mode. In the same study, there was shown a positive correlation between older age an d calling. In addition the Kennedy and Wellman (2007) study found that older females tried to avoid using text
27 based communication. Communication research concerning age has examined the age of the participant For the current study, the age of the partne r was examined, in w hich the age of the participant r esults provide a framework with which to base this hypothesis on. The second hypothesis expected females to have a higher frequency o f both calling and texting as comp ared to males. This hypothesis wa s based on results from Pierce (2009) who observed a significant gender difference, where girls used cell phones for both calling and for texting more than boys. Raacke and Bonds Raacke (20 11) also documented females texting and forwarding texts at higher rates than males which bolsters this hypothesis. For final supportive results pertaining to texting, Ishii (2006) found a significant, positive effect for gender, which indicated a strong a ssociation between females and the use of text messaging. There were counter findings in the Underwood et al. (2012) study showing equal communication rates of males and females, but the bulk of the research available consistently shows a trend of females communicating more than males. Given the findings of these other researchers, it is reasonable to expect that the call and text frequencies will be higher for females than those for males. The third hypothesis stated that calling would be used most with pa rents than any other close relationship partner Findings that the positive and beneficial relationship qualities of greater satisfaction, intimacy, support, and instrumental aid were related to higher frequencies of college students calling parents lend g reat support to this expectation (Gentzler et al., 2010). Additionally, laying a solid foundation for such expectations, a nationwide Pew Research study of the United States mentioned how the
28 teenagers with cell standpoint, mothers were reported to call their kids more than fathers (Kennedy & Wellman, 2007). This provides more evidence for possible amplifi ed frequencies of the calling function being used for parent child communication than any other dyad types. There were no hypotheses made in regards to siblings, intimate partners, and best friends because it is felt that more research needs to be performe d concerning these partner types before confident hypotheses can be drawn The fourth and final hypothesis expected that actual habits, based on cell phone record logs, and perceived habits as collected from surveys, would be correlated for information pertaining to a 7 day period. A recency bias, where memories are biased toward recent events, was a factor that needed to be controlled for so that accurate information could be collected. Providing an effective way to take the recency bias into account, the work of Eagle, Pentland, and Lazer (2009) was adopted. Their research found day time frame. This effect indicates a degradation of memory for events develops around approximately 1 week. Supported by this research a 7 day period of time was selected. In other supportive literature, the recency of the events that were self reported in Underwood et al. (2011) is not stated, however their study did report a positive correlation between billing records and self reported cell phone use. Note, this correlation was w eak, so more research is required to fully substantiate the notion of accurate self
29 week, which may have accounted for the weak correlation and so this study sought to obta in data which might clarify this finding. Method Participants Forty undergraduate students [19 males, 21 females] between the ages of 18 through 28 years old [ M =20, SD = 1.895], from a small public liberal arts college in South west Florida participated. A ll of the participants had a living family member (parent or sibling), had someone he/she considered an intimate partner, had someone he/she considered a best friend, had a smartphone, and had unlimited texting and calling. All partner types were mutually exclusive. The family members, intimate partner, and best friend were all required to have a cell phone with texting capabilities. Participants were recruited using convenience sampling techniques by sending out an email to a student body email list or wer e approached in public areas on campus to participate in a study regarding communication habits. All participants met with the researcher alone to complete the initial survey and then the calling and texting records collection. Materials A computer was use d by the participants to take the SurveyMonkey survey (see Appendix A) and then used by the experimenter to collect call and text logs. Survey Measures Communication Behaviors Measure One item in the survey was used to assess the perceived main function of the smartphone by asking the participant, out of the calling function or the texting function, which he/she believes is personally utilized more. In the survey, for each of the close relationship types (parent, sibling if applicable,
30 intimate partner, and best friend), the self rep orted typical frequency of calling communication was asked using a modified version of a communication frequency scale developed by Gentzler et al. (2011). The current study used an 8 point scale to find the frequency of each type of communication within a one week time period (1=Less than once a week, 2= Once a week, 3=Once every other day, 4=Once a day, 5=Every few hours, 6=Every other hour, 7=Every hour, 8 =Several times each hour), rather than the original scale that measured frequency over a year. Another change was that each relationship type was specified for the frequency reports instead of a general frequency to any person. In the survey, for each of t he close relationship types (parent, sibling if applicable, intimate partner, and best friend), the self rep orted typical frequency of texting communication was asked using the same 1 week communication frequency scale. Reasons why Text instead of Call/Cal l instead of Text In the survey, the reason ( s ) why someone chooses to use one communication mode (call or text) over the alternative mode was gathered. There were eight of these questions phrased in the format ur (close partner) instead of (alternate A coding scheme was created for these items by adapting codes from Mandell and Muncer (2007), Kamibeppu and Sugiura (2005), Gentzler et al. (2011), Ishii (2006), Pierce (2009), Raacke and Bonds Raacke (2011), and Shadyad et al. (2011). Sixteen final codes were selected. The individual requirements for a code to be assignedare presented below.
31 1. Sending a Short/Non urgent Message : Any mention of not having a lot to say, or communicating a short i dea, brevity or quick (implying short) message, or any suggestions of minimal information being exchanged, or something not very 2. Lengthy Conversations : Any mention of n eeding/taking up a lot of time, having a lot to 3. Immediate Contact Required : Any mention of urgency, or things being time dependent, or limited by time, or the desire to know a s soon as possible. 4. Personal Ability : Any mention of social anxiety or low social skills or allusion to being better at communicating through certain media, or not being able to articulate as well with other media, or how other media makes them nervous or it is difficult to communicate because of intoxication. 5. Comfort/Right Amount of Closeness : Any mention of the mode being the most comfortable/preferable to use, or being the preferred mode because of the degree (high or lack) of intimacy it involve s, or right feel/dynamic of interactions. 6. : Any mention of the partner not being able or apt at using others modes of communication, or how the partner is best at using a certain mode of communication. 7. Covert Communication : Any mention of either partner not being able to use other modes of communication at the time (busy), or the situation making it so the mode of communication is the only appropriate way to convey the information, or not wanting to disturb or bother the partner (work or sleep), or about noise being an issue that makes
32 text them), or different schedules or to read at later time. 8. Multiple People at Once : Any mention of h ow several people besides the partner can be contacted at the same time, or how the person is not limited to communicating with only that partner at the time. 9. Convenience/Ease : Any mention of how the mode is fast/easy to use, is convenient or efficient to use (saves time/energy), or that it is readily available to use, or does not take much effort to use, or implied convenience because other mode not, or how things are easier such as expressing oneself. 10. Time to Think : Any mention of being given more time, or not being rushed, or the 11. Get/Give Information (practical) : Any mention of wanting to give or receive practical information such as how to do something (e.g. change a tire), where to go for something (e.g find mechanic), or anything that is intended to assist in the efficiency of daily tasks, oriented) exchange of info, or answering a question (any mention of a question th at does not mention a social aspect), or to solve a problem/figure something out. 12. Get/Give Information (social) : Any mention of wanting to give or receive social information which could be about themselves or other people or of other interpersonal rela lengthy conversation), or any emotional information/feelings about themselves or other being/status, or about how the day went (e.g. tell a story).
33 13. Avoid Confrontation/Emotional Situation : Any mention of trying to avoid any disagreements, or avoiding any emotional or emotion based conversations, or avoid charged/heated conversations, or touchy subjec ts. 14. Avoid Long/Verbal Conversation conversation, not wanting to be stuck talking, or does not need to be long (participa nt does not want partner communication to be long). 15. Economical/Financial : Any mention of money, costs, or the fiscal value of the mode. 16. Other/Reciprocation : Anything that does not fit in other categories, or is not a real reason why use (e.g. calli ng them back), or say that is the only way they communicate or they always communicate, or involves non text communication (e.g. picture messages), or is a vague statement (e.g. to talk), or is only about another mode. After all answers were coded by the p rimary researcher, a second researcher was recruited to do a reliability check of 25% of the response units coded. A random number to code. Individual units were define d as the totality of one open ended response for each question asked. Multiple codes could be assigned per unit. Each code was either considered present or absent. To compare coding decisions, the total number of codes assigned per unit and the congruency of which codes were assigned was calculated. From these numbers the All or none Agreement Percentage and the Partial Agreement Percentage values were found. For the All or none Agreement value, both researchers had to agree completely where each assigned the exact same codes per response, regardless of how many were listed, providing an all or none agreement. For the Partial
34 value, a count was taken of the number of individual c odes assigned, noting the matches, and the mismatches. To be counted as a mismatch there were either different codes assigned, or if the researchers counted a different number of codes (e.g. Researcher A assigned 3 codes to a unit and Researcher B assigned 4 codes to the same unit; assuming the 3 were the same, the lack of a code/extra code was a mismatch). Whe n the code s were compared, the All or none Agreement Percentage came out to 81% agreement and the Partial Agreement Percentage was 88%. Demographics The demographics gathered were the age and biological sex of the participant and all of his/her close relationship partners. In addition, the estimated geographical distance between the participant and each close relationship individual was recorded. A 1 5 scale was used for measuring distance where 1= Within walking distance 2= Same city 3= Same state 4= Same country 5= On different countries/continents. Call and Text Logs The call and text log information was taken from the participant by having him/her take out his/her smartphone and click the call and text icons to look up the recent phone records for the past 7 days. All of the calls and texts made to each of the close relationship partners were recorded, noting which partner was being contacted, who i nitiated the contact, and the date and time of the call/text. Attempted but missed calls were also recorded. Procedure A recruitment email was sent out to all of the student body asking for participation in a psychology study focusing on communication habits and listing the requirements of being a student who was 18 years or older with a smartphone that has
35 unlimited calls and texting. Also, the participants needed to have at least one person for each relationship partner (family, best friend and intimate partner) who also had a cell phone so texting was a possibility. In the email, each participant was asked to schedule an appointment with the experimenter through an online appointment manager. When an appointment was made, the experimenter emailed the participant instructing him/her not to delete call or text information for the 7 days prior to the appointment. The appointments took place in a quiet study room. At the appointment the participant was verbally questioned on the requirements to make sure he/she qualified and then asked to provide written consent. After he/she gave consent, then he/she was asked to u se a provided computer to fill out a SurveyMonkey survey (See Appendix A). The survey instructions specified that the participant was to answer the questions with reference to the individual who they most often communicated with for each relationship partn er (parent, sibling (if applicable), best friend, and intimate partner), (e.g. most often The participant was told to use the same selected relationship individuals for the subsequent call and text log data co llection as they chose for the survey. Once this was complete, the researcher took the computer back and asked the participant to take out his/her smartphone. From there, the participant was once again asked to think of the same person from the survey whom he/she most communicated for each of the four close relationship partners (family parent, family sibling, intimate partner, best friend). Then the participant was asked to go into the call and text logs. The researcher started with the call logs, and had the participant verbally report all of the calls made between the participant and each selected close relationship partner (parent &/or
36 sibling, intimate partner and best friend) in the past 7 days. The verbal ly reported data were typed into computer files The information collected for each contact was: which partner was the contact, who initiated the contact, the date, and time of the call. After that, using the same verbal reporting methods, the participant was asked to go into his/her text logs between him/her and the same close relationship individuals for the prior 7 days recording the same categories as calls. Only the participant looked at the phone so no personal information or content was known to the experimenter.After all of the data were collect ed, the participant was debriefed and then given his/her $5 payment for participation. Results Descriptive Results of Coded Reasons for Communication Choice ended answers given when asked why they might call in stead of text, or text instead of call when communicating with each of the four close relationship partners Since the Other/Reciprocation code was used to gather the answers that were not actually reasons, but were a choice based on something else (e.g. re ciprocation) or was an extraneous comment, the number of code occurrences for this code will not be part of the reported findings. In this way a more accurate depiction of the main reasons participants gave will not be skewed in ranking order. The frequenc y distribution of codes assigned by close relations hip partner by mode is found in Table 1. All codes could have a maximum of 40 for all questions except the ones pertaining to siblings, which had a potential maximum of 34 as 6 participants did not have si blings.
37 To see a clear picture of the how participants differed in their reasoning by partner type, the top four reason categories were identified, as long as the code had at least four participants reference it. The four most frequently cited reasons show a strong indication that the choice to call or text is primarily based on the capabilities afforded by the mode, such as covertness for texts and lengthy conversations for calls When the four main reasons were found for each individual partner question, all of the calling reasons were examined together. When ranks were collapsed across partners for calling, the main reasons mentioned were Immediate Contact Required, Lengthy Conversations and Get/Give Information (practical) and Get/Give Informa tion (social). The same process was used to find the four most cited reasons for texting. The most often cited reasons when texting was collapsed were Covert Communication, Sending a Short/Non urgent Message, Convenience/Ease, Get/Give Information (practic al), and Avoid Long/Verbal Conversation. From a visual inspection of each of these in Table 1 it is clear that there are no obvious differences in reasons for communication based on which partner was being contacted. Actual vs Perceived Frequencies Pearso n correlational analyses were run comparing the actual frequencies of calls made to each close relationship partner by the frequencies the participants self reported on the survey. The same analysis was then run for actual and perceived frequencies of text s. Of all the Pearson correlation tests run there were only two correlations worth mentioning based on their significance levels and coincidentally they both pertain to
38 interactions with siblings. When the participant's call logs and the perceived call fr equencies with their siblings were tested it was seen that there was a positive correlation between the actual and perceived calling frequencies with siblings, r (31)= 0.61, p < .001. As well, not significant, but approaching significance were the actual an d perceived texting frequencies with siblings, r (31)= 0.31, p = .07. For a ll remaining correlations between actual and perceived contacts see Table 2 and all other cor relations were not significant ( all p ) Age Correlations Pearson correlational analyses were run comparing the actual frequencies of calling communication with each close relationship partner by the age of each close relationship partner. The same analysis was run with texts. All the Pearson correlation tests ru n across all partners by both modes were not found to be significant except for the data pertaining to communication with parents. revealed that there was a positiv e correlation, r (38)= 0.325, p < .05. See Table 3 for all other correlation values ( all p ) Distance Correlations Correlational analyses were run comparing the actual frequencies of calling communication with each close relationship partner by the p hysical distance between each close relationship partner. The same analysis was run on texts. None of the Pearson correlation tests run across all partners and both modes were able to find significance between the physical distance apart and the calling a nd texting behaviors. All correlation values can be found in Table 4 ( all p )
39 Gender Frequencies: Two way ANOVA In order to determine whether there was a relationship between gender of participant and mode, a 2 (gender: male/female) x 2 (mode: call/ text 3 ) two way mixed ANOVA with repeated measures on the 2 nd factor was conducted. There were two outlying data points, one for calling and one for texting, so analyses were run on both the data with and without the outliers.The dependent variable was number of calls/texts for both data sets The full data set, including the ou tliers is presented first. T here was no main effect of mode, as calls ( M =5.105, SE = .858) and texts ( M = 6.759, SE = 1.18) were equally frequent F (1, 36 ) = 1.56, p =.219. There was no main effect of gender (males: M = 5.579, SE = 1.148, females: M = 6.29, SE = 1.092), F (1, 36 ) = .199, p =.658, and no significant interaction between gender and mode(see Figure 1), F (1, 36 ) = .481, p = .492 This indicates that gender did not predict rates of communicat ion The same 2 (gender: male/female) x 2 (mode: call/text) ,mixed ANOVA with repeated measures on the 2 nd factor was conducted using data without the outliers.There was no main effect of mode, with calls ( M =4.531, SE = .717) and texts ( M = 5.867, SE = .828) being equally frequent F (1, 30 ) = 2.005, p =.1 65. There was no main effect of gender (males: M = 4.972, SE = .891, females: M = 5.425, SE = .845), F (1, 30 ) = .136, p =.714, and no significant interaction between gender and mode (see Figure 2 ), F (1, 30 ) = .167, p = .685. This ind icates that when lookin g at more representative rates of communicati on gender was still not predictive of call/text rates 3 These analyses were run on the data about communication coming from the participant to the partner and not from both interlocutors.
40 Mode Frequencies: Three way ANOVA 4 To see if there was a relationship between initiator and mode and partner type, a 2 (initiator: participant/partner) x 2 (mode: call/text) x 4 (partner type: parent/sibling/intimate partner/best friend) three way repeated measure s ANOVA was conducted. The dependent variable was number of calls/texts See Figure 3 for the means of partner type where t here was a main effect of partner type, such that there was a significant difference between the number of contacts with parent s sibling s intimate partner s and best friend s F (3, 31 ) = 6.823 p < .000 From a visual inspection, it seems that intimate partners significantly differed from all the other partner types ( M = 2. 617), and this was especially evident for when intimate partners texted ( M = 3.469) (see Figure 6 ). This suggests that intimate partners texted at higher rates than all the other partners. There was also a main effect of mode, such that there were significant ly fewer calls than texts (call: M =1. 188 SE = .1 93 text: M = 2.020 SE = .2 86 ), F (1, 3 1 ) = 7.258 p =.011 There was no main effect of initiator as partic ipants( M = 1.598 SE = 218) and partners( M = 1. 609 SE = .190 ) initiated contacts equally often F (1, 3 1 ) = .005 p =.942 None of the two way interactions were significant (initiator by partner F (3, 3 1 )= 807 p =. 493, initiator by mode F (1, 3 1 ) =1. 499 p = 230,mode by partner F (3, 31 )= 1.5 63 p = .2 04) The means for these analyses are depicted in Figure 4, 5, and 6 respectively Finally, there was a near significant interaction between initiator, mode, and partner type, F (3, 3 2 )= 2.426 p =.0 71 See Figure 7 for a graph of the individual means for the interaction between initiator and partner type for calling. See Figure 8 for a graph 4 These data were positively skewed but transformations (Ln) did not fix ske wedness, thus the ANOVA was used as a default because it is robust.
41 of the individual means for the interaction between initiator and partner type for texting. From inspecting the t wo graphs it appears that the source of the interaction is in communication with parents, where calls were initiated by participants but texts were more often initiated by parents. The other partner types did not show this switching of initiation roles acr oss calls and texts. The two and three way a nalyses in the first sample had varying N values because of issues with whether a participant had a sibling or not, so another analysis was conducted without siblings. A 2 (initiator: participant/pa rtner) x 2 (mo de: call/text) x 3 (partner type: parent /intimate partner/best friend) three way repeated measures ANOVA excluding siblings was performed for all 40 participants. There was no main effect of partner type, such that there was no significant difference betwe en the number of contacts with parent s intimate partner s and best friend s F (2, 39 ) = 2.096 p =.130, (see Figure 9). There was a main eff ect of mode, such that there were significant lyfewer calls ( M =1. 450 SE = 220) than texts ( M = 2.154 SE = 304 ), F (1, 39 ) = 4.353 p =.0 44 There w as no main effect of initiator, as participant s( M =1. 825 SE = 248), and partners( M = 1. 779 SE = .18 6 ) initiated contacts equal lyoften F (1, 39 ) = .083 p =.774 The initiator by partner type interaction was not significant, F (2 3 9) = 1. 041 p =. 358 The initiator by mode interaction was not significant, F (1, 3 9 ) = 2.037 p = 162 However, the mode by partner type interaction was approaching significance F (2, 39 )= 2.616 p = 079, which, from visual analysis seems to have occurred for intimate partners texting ( M = 2.838). The means for these analyses are depicted in Figures 10, 11, and 12 respectively
42 Finally,there was no significant interaction between initiator, mode, and partner type, F (2 3 9 )= 2.122 p =. 127.See Figure 13 for a graph of the individual means for the interaction between initiator and partne r type for calling. See Figure 14 for a graph of the individual means for the interaction between initiator and partner type for texting. Although some p attern of calls to parentsvs texts from parents is seen in the means, this did not reach significant levels. Beyond these results, all the rest of the results show that there was not a significant difference in the calling and texting habits between who st arted the contact, as well as no real difference in the communication habits between parents, siblings, and best friends, and no significant difference in the overall calls and texts performed. The reasons for communication mode were shown to have a very c lear split between the two modes, with very little overlap between them. Beyond the top four reasons cited for each questio n, there were a few participant s who mentioned some other reasons but they appeared in very low quantities. Discussion The objective of this study was to investigate text are predicted by communication partner type, age and sex demographics, andreasons for mode choice In addition to this purpose, a review of the validity of self reports was also examined by comparing actual communication records to corresponding self percep tions of communication habits. As predicted, t he age of parents was shown to positive ly be correlated with calling. Contrary to predictions, f emales and males were fou nd to have relatively equal frequencies of both calling and texting. Texting was shown to be used more than calling ,where collective findings that approached significance
43 suggested that texting was used most by intimate partners. In opposition to predictio ns, when comparing actual behaviors based on cell phone record logs and perceived habits collected from surveys during to a 7 day period, overall no relationship was found between the data sets of perceived habits and actual behavior. These results were pa rtially supported by previous literature. Partner Type P articipants had higher communication rates with intimate partners than other partners. This was not in line with the Lenhart et al. (2010) data which showed best friends produce the highest communica tion frequency. The current study forced participants to pick different people for each partner type. There may have been an overlap effect where intimate partners are sometimes thought of as best friends, accounting for this discrepancy between results. I n reviewing past literature, it is unclear how other studies approached this methodological issue. Because intimate partners often times share resources, such as cars, food, and money, there is a more practical need to be tner to manage these concerns. T here was not a significant interaction of partner type and modewith siblings included ; however, texting between intimate partners also had a very large mean value (see Figure 6) The results together find heavy texting communication occurring between college students and their intimate partners. This was given more support from the nearly significant interaction found between partner type and mode from the three way ANOVA run wit hout siblings (see Figure 1 2 ). This suggested trend of intimate partners texting the mo st is in opposition to Coyne et al. (2011) whose results indicated that calling was the primary mode used between intimate partners however, it should be noted that tex ting
44 was shown to be a close second in mode use Also, intimate partners showing the highest communication pattern also differs from the results of Lenhart et al. (2010) results that friends were shown to be texted the most as mentioned above Beyond poss ible overlap discrepancies, these differences may be due to the fact that the Coyne et al. study used a wider age range, including many older participants, which younger age has shown to be correlated with a higher propensity to text. Also, the Lenhart et al. study had the opposite effect, where the age group consisted of only teenagers. This is relevant because it has been shown that developmentally, this age group is more concerned with friends. Whereas, early adulthood, which the current participants wer e in, is seen as a time when people start to shift into a more romantically preoccupied stage of life, where intimate partners may hold equal or priority status relative to friends. Mode More of the communication was conducted via texting rather than call ing. The more often the chosen communication mode than calling. This finding is also complemented by the work of Kamibeppu and Sugiura (2005), which saw higher rates of texting as opposed to calling, and the work of Skierkowski and Wood (2012), which specifically focused on a college aged sample population, and also found that texting was the main mode of communication used. From these studies, this pattern is seen to hold over eight + years, supporting that this is a stable trend in communication behavior. they might text instead of call, the reasons of Covert Communication, Sending a Short/Non urgent Message, Convenience/Ease, Get/Give Information (practical), and
45 Avoid Long/Verbal Conversation arose most providing some conceptual understand of why this popular texting pattern has not show n to wav er over the years. Especially since the sample population is composed of students, there is a necessity to be covert if they want to communicate during a class, which occupies a large portion of their time. As well, texting being regarded as conven ient may be a result of wanting to convey a message to someone without needing to stop working on homework assignments, whereas calling necessitates postponement of current activities to be able to devote attention to what one is saying and hearing. Intera ctions There were no significant interactions found between the variables of mode, partner type, and initiator for the analysis with siblings however the overall three way ANOVA did approach significance. It seems that the near significant result lies with the communication between the participants and their parents. I t was seen that participants called the parents more and the parents texted the participants more.This flip in main mode used between the pa rticipant and their partner is not seen between any of the other partner type dyads. As well, the three way interaction w ith o ut s iblings was not significant, but there the same pattern is seen, although it was not large enough to reach sig nificance. This pattern is very surprising since the literature of Raacke and Bonds Raacke (2011) and Skierkowski and Wood (2012) showed evidence that texting is primarily used by a younger demographic. As well, the results found by Jin and Park (2010) an d Ishii (2006) indicated that age was correlated with calling, which parents are inherently much older than the participants, so intuitively one would think they would call more. I n light of this finding, a possible explanation might be that the student
46 pa preferred mode of communication and subsequently show an opposite mode choice effect. Also, a possible explanation might be that the participants chose to call because they needed to explain a complex situation or to ask for something, such as explaining how he/she got into a car accident and asking for money. On the other hand, parents might have chosen to text because their message was simple and brief, such as confirming that he/she had put Age The age of the partner was tested for correlations with calling and texting behaviors. Partially supporting the hypothesis, a neffect of mode was found based on partner age. When mode was observed by partner type, there was a positive correlation between the age of the parent and rates of calls. T he older the age of t he parents, the more calls the parents/participants made to /received from each other So, as the age of a parent increases, so does the likeliho od that she or he will be in communication with a child by calling. This finding agrees with the research of Ishii (2006) who also found a positive correlation between older age and making calls. As mentioned previously, this study did differ from reviewed literature by examining the age of the person contacted instead of the participant. However, this fact seems to only expand the ways of studying communication behavior by illuminating that both interlocutors are applicable to the research. This age relate d relationship with calling might arise because parents were the oldest age group examined. Since they are older they have grown up with the ability to make phone calls for the entirety of their life, whereas texting is a very recent development. Older par ents may be more inclined to make calls as a force of habit, as
47 well as, they may be being more comfortable with calling because they have been doing it for so long. Gender From the analyses o f participant gender and mode both of the main effects as well as the interaction s with siblings were not significant. This is an interesting finding since it means that in the present sample females and males did not differ in calling and texting habits. However, this finding goes against most of the literature, such as Raacke and Bonds Raacke (2011), who found females to text at higher rates than males, or the findings that females use both calls and texts more frequently than males (Pierce, 2009). Differences in results may be attributable to difference is methodology since the current study incorporated actual logs but the studies mentioned above used only questionnaires 012) finding of equal communication rates between genders.Another possibility is that this gender difference is shifting historically. Physical Distance The physical distance between the participant and each partner type was conducted as a control measure to make sure any texting or calling habits were not influenced by this factor. For example, it is plausible that low calling rates might be a result of communicators being in such close proximity that they do not need to call each other. In the analysis, t frequency of communication being a consequence of distance, thus ruling out this explanation.
48 Communication Mode Reasons In an examination of the reasons provided by participants for why they might choose to call vs. text or text vs. call a partner, there were no consistent differences seen for contacting a particular partner Nonetheless, the data did appear to show patterns of reasons between calling and texting scenarios overall. When looking at the main reasons given for calling, there seems to be a theme of very synchronous communication being sought, such as seen by the two most referenced codes of Immediate Contact Required, and Lengthy Conversations. As for texting, the asynchronous aspect of the mode seemed to be highlighted by the codes of Convenience/Ease, and Avoid Long/Verbal Conversation. When it comes to mode cho ice, the synchronicity of the medium seems to be a contributing factor in deciding which mode to use. This is in agreement with for their communication mode choice. synchronicity of the mode to be a factor in their choice of how to communicate. Actual vs Perceived Communication Actual calling and texting logs were compared to the perceptions of the communication behaviors that occur between the participant and each close relationship partner type. The current study used a one week interval, which Eagle et al. (2009) argued was reliable for memory reports of frequency judgments. The only significant correlations were found with siblings, where the calling records were highly positively correlated with the self reported data. This means that the perceptions of communication were highly attuned to the real frequency of communication for only this type of
49 communication partner. The texting logs with siblings were near significance when run against the perceptions of text messaging, suggesting that the perceptions of texting with siblings are almost accu rate depictions of texting behaviors. However, these results being found specifically only for siblings would seem to mean that there is something special about siblings. From a visual examination of the means, it is apparent that communication with siblin gs was the lowest compared to the other partners. Consistently, participants would report 0 calls and 0 texts in the week time frame regarding communication with siblings. The calling frequencies with siblings ranged from 0 7, and the texting frequencies w ith siblings ranged from 0 6. It is believe d that this lack of communication is the cause for these results since it is easier to keep track of lower rates of contact Whereas, the ranges of calls and texts of the other three partners were much wider, such that calls with parents range d from 0 16, calls with intimate partner ranged from 0 30, calls with best friends ranged from 0 10, texts with parents ranged from 0 12, texts with intimate partners ranged from 0 45, and the texts with best friends ranged from 0 19, all of which have rate s that maybe m ore difficult to keep track of. The s e data and the rest of the non significant findings with the other close relationship partners suggest there maybe reason to doubt self report data, even using a 1 week time period. This calls for more resea rch to be conducted concerning this accuracy problem in communication research. This may i ndicate a need for shorter than 1 week intervals of study to stack against self reported data. The technology of smart phones now makes this an easily realizable way to measure behaviors.
50 Limitations The se conclusions of a ccuracy however must be made cautiously due to limits in the way perceived calls and texts were measured Measuring the frequency of calls and texts in the survey portion of the study used a 1 8 scale, but the actual call and text frequencies were composed of the total number of calls/texts made within the 1 week period. This may not have been an ideal way to assess correlations of accuracy because the two different measures did not allow for a one to one comparison. This difference may have caused discrepancies in perceptions, leading to differences in accuracy testing. In terms of measuring frequency, perhaps asking participants how many total calls/texts they think they make in a week might provide a more fitting measure to compare against actual records. Additionally, using a shorter time frame, such as one day, might make perception records more in line with actual behaviors. The questions of where does the accuracy hold and how far can it be reliable are serious inquiries that need to be found. If a 1 day time frame works but a 1 week time frame does not, then this has implications for what the field of communication has as existing data in survey studies. The analysis of the actual vs per ceived frequency of communication may have been initially error prone due to unclarified wording of the perceived frequency questions in the survey. When the actual logs were recorded, the participant was asked to only provide information on initiation of whole conversations instead of each individual reply in a conversation. This conversation based count was not specified in the questions asking for perceptions of communication which may have le d participants to provide communication data of two different scales of measure. This conflict in measures was not reali zed until after all the data were collected and analyzed. For future studies, it is
51 recommended that the researchers clearly state that only whole conversations need to be reported. Whole conversati ons and not every reply in conversation is suggested because it gets at the heart of the communication behavior without becoming muddled in an Future Directions Future research into how to collect accurate communication data is desperately needed in the field. At the moment, there is no standardized method for collecting information on communication habits. Few log data collections and many survey/questionnaire collections are being perf ormed, with many using self created scales. If scales are to be used, there needs to be a standard scale and a standard time frame for use so there is consistency among the results, aiding in the reliability of reported communication research. If both surv ey and log data are collected, there needs to be consistency of communication measures between them. As mentioned above, a variety of time intervals, in cluding a week long time period, should be tested against communication perceptions in future research to work on finding an ideal time span from which mostly accurate survey data can be collected. Conclusion Communication behaviors are changing with the changes in technological developm ent. Yet, some aspects of communication like who we speak to and who we are stay the same. Between these changeing and stable elements, a dynamic process happens leading us to make the final decision to call or text. The present study examined how partne r type, mode type, and demographics all play a role in communication frequency, and found intimate partners to receive the highest rates of communication by
52 college students. College students also were shown to use texting at higher rates than calling How ever, these data suggest calls and texts may vary when communicating with parents Altogether, these findings provide implications about texting and intimate partners for further research to have a better understanding of the factors that spur more communi cation between people. In time, further communication research can be shaped around these new understandings, possibly leading to ways of enhancing communication modes so the relationships involved may be enhanced as well.
53 References Auter P. J. (2007). Portable social groups: willingness to communicate, interpersonal communication gratifications, and cell phone use among young adults. International Journal of Mobile Communications, 5 ,139 56.Retreived from http://www.biomedsearch.com/sci/Portable social groups willingness to/0004401059.html Baym, N. K., Zhang, Y., Kunkel, A., Ledbetter, A., & Lin, M. (2007). Relational quality and media use in interpersonal relationships. New Media & Society 9 (5), 735 75 2.doi:10.1177/1461444807080339 A descriptive study of the media use of individuals in romantic relationships. Family Relations: An Interdisciplinary Journal Of Appl ied Family Studies 60 (2), 150 162. doi:10.1111/j.1741 3729.2010.00639.x DeGeneres, E. (2011). Seriously -I'm kidding New York: Grand Central Pub. Eagle, N., Pentland, A., &Lazer, D. (2009).Inferring friendship network structure by using mobile phone d ata. PNAS Proceedings Of The National Academy Of Sciences Of The United States Of America 106 (36), 15274 15278. doi:10.1073/pnas.0900282106 use of electronic communicatio n with parents: links to loneliness, attachment, and relationship quality. Cyberpsychology behavior and social networking 14 (1 2), 71 74. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/20973676
54 Goby (2011). Psychological underpinnings of intrafamilial computer mediated communication: A preliminary exploration of CMC uptake with parents and siblings. CyberPsychology, Behavior, and Social Networking, 14, 365 370 Hamburger, Y. A., & Ben Artzi, E. E. (2000).The relationship between extraver sion and neuroticism and the different uses of the Internet. Computers In Human Behavior 16 (4), 441 449. doi:10.1016/S0747 5632(00)00017 0 Ishii, K. (2006). Implications of Mobility: The Uses of Personal Communication Media in Everyday Life. Journal Of Communication 56 (2), 346 365. doi:10.1111/j.1460 2466.2006.00023.x Jin, B., & Park, N. (2010). In person contact begets calling and texting: Interpersonal motives for cell phone use, face to face interaction, and loneliness. Cyberpsychology, Behavior, And Social Networking 13 (6), 611 618. doi:10.1089/cyber.2009.0314 Kamibeppu, K., &Sugiura, H. (2005). Impact of the mobile phone on junior high school CyberPsychology& Behavior 8 121 130 Kenned y, T., & Wellman, B. (2007).The networked household. Information, Communication & Society 10 (5), 645 670. doi:10.1080/13691180701658012 Lenhart, A., Ling, R., Campbell, S., & Purcell, K. (2010).Teens and mobile phones. Retrieved from http://pewinternetorg /Reports/2010/Teens and Mobile Phones.aspx
55 Lenhart, A., Madden, M., &Hitlin, P. (2005). Youth are leading the transition to a fully wired and mobile nation. Pew Internet & American Life Project.Retrieved on April 16, 2007. Available from: http://www.perinternet.org/ Ling, R., &Stald, G. (2010). Mobile communities: Are we talking about a village, a clan, or a small group?. American Behavioral Scientist 53 (8), 1133 1147. doi:10.1177/0002764209356245 Madell, D. E., &Muncer, S. J. (2007). Con trol over social interactions: An important reason for young people's use of the Internet and mobile phones for communication?. Cyberpsychology& Behavior 10 (1), 137 140. doi:10.1089/cpb.2006.9980 Miyata, K. (2006). Longitudinal effects of mobile interne t use on social n etwork i n japan Paper presented at the annual meeting of the International Communication Association, Dresden International Congress Centre, Dresden, GermanyOnline
56 Raacke, J ., & Bonds Raacke, J. (2011).An investigation of the dimensions of SMS communication use by college students. Individual Differences Research 9 (4), 210 218. Shahyad, S., Pakdaman, S., Hiedary, M., HasanZadeh, M., MirnaderMiri, Asadi, M., & ShirAlipour, A (2011).A study of psychometric characteristics of the questionnaire on the type of usage of S.M.S (motivation, content, frequency). Procedia Social and Behavioral ciences, 15 (0), 932 936. doi: 10.1016/j.sbspro.2011.03.21 S Skierkowski, D., & Wood, R. M. (2012).To text or not to text?The importance of text messaging among college aged youth. Computers In Human Behavior 28 (2), 744 756. doi:10.1016/j.chb.2011.11.023 Underwood, M., Rosen, L., More, D., Ehrenreich, S., &Gentsch, J. (2012). The BlackBerry project: Capturing the content of adolescents' text messaging. Developmental Psychology, 48 (2), 295 302. doi: 10.1037/a002591
57 T able 1 Reason Code Frequency Distribution by Partner and Mode Partner by Mode Reason Parent Sibling Intimate partner Best friend Call Text Call Text Call Text Call Text Lengthy Conversations 17 0 12 0 13 2 17 1 Covert Communication 0 17 0 12 0 17 0 14 Immediate Contact Required 14 2 8 0 8 1 11 0 Sending a Short/Non urgent Message 0 11 0 7 0 11 0 8 Convenience/Ease 8 9 1 6 4 11 4 12 Get/Give Information (practical) 7 5 7 5 11 8 10 7 Get/Give Information (social) 2 2 7 2 9 3 7 6 Avoid Long/Verbal Conversation 0 10 0 7 0 3 0 3 Comfort/Right Amount of Closeness 1 2 1 2 0 0 0 2 Economical/Financial 0 1 0 2 0 1 0 1 Avoid Confrontation/Emotional Situation 1 0 0 0 0 0 0 0 Personal Ability 0 3 0 0 0 1 0 1 0 1 0 0 0 0 0 0 Multiple People at Once 0 0 0 0 0 0 0 0
58 Time to Think 0 0 0 0 0 0 0 0 Other/Reciprocation 15 4 9 8 17 8 8 7 Max Possible 40 40 34 34 40 40 40 40
59 Table 2 Pearson Correlations of Partner by Mode for Actual vs Perceived Frequencies Partner Mode Parent Sibling Intimate Partner Best Friend Calling .09 .61*** .17 .11 Texting .07 .31 .18 .06 Note. All degrees of freedom are out of 38, except Sibling are out of 29. *** p <.001
60 Table 3 Pearson Correlations of Actual Calling/Texting Frequencies by Partner's Age Partner Mode Parent Sibling Intimate Partner Best Friend Calling .33 .15 .04 .15 Texting .16 .18 .09 .06 p < .05
61 Table 4 Pearson Correlations of Actual Calling/Texting Frequencies by Partner's Physical Distance Partner Mode Parent Sibling Intimate Partner Best Friend Calling .13 .17 .09 .26 Texting .03 .00 .08 .05
62 Figure 1 Calling and Texting Frequencies by Gender (with outliers)
63 Figure 2 Calling and Texting Frequencies by Gender (without outliers)
64 Figure 3 Communication Frequencies by Partner Type
65 Figure 4 Communication Initiation Frequencies across Partner Type
66 Figure 5 Communication Initiation Frequencies by Mode
67 Figure 6 Mode Frequencies by Partner Type
68 Figure 7 Calling Frequencies by Partner Type and Initiator
69 Figure 8 Texting Frequencies by Partner Type and Initiator
70 Figure 9 Communication Frequencies by Partner Type (without Siblings)
71 Figure 10 Communication Initiation Frequencies across Partner Type (without Siblings)
72 Figure 11 Communication Initiation Frequencies by Mode (without siblings)
73 Figure 12 Mode Frequencies by Partner Type (without Siblings)
74 Figure 13 C alling Frequencies by Initiator across Partner Type (without Siblings)
75 Figure 14 Text ing Frequencies by Initiator across Partner Type (without Siblings)
76 Appendix Appendix A