Kin Relationships and Email: Bringing Families Closer Together

Amy Epner and Kevin H. Gross

Abstract

Using data from the PEW Internet and American Life Project, March 2000 Survey, this study investigates how using computer-mediated communication (i.e., email) to communicate with family members affects family relationships. First, we look at who uses email to contact family members. Then we look at whether those individuals using email perceive that email affects their family relationships. Results show that younger Caucasian females with higher incomes living in two-adult households are more likely to email family members; using email is associated with increased family communication. Overall, emailing family members is thought to benefit family relations. Findings are consistent with the knowledge gap hypothesis; however, a displacement model of computer-mediated communication is not supported.

Introduction

According to the U.S. Census Bureau, approximately 6% of family households with children moved to a different county, state, region, or abroad in 2002 (U. S. Census Bureau, 2003). For some of these families, this means they have moved away from their family of origin, leaving them geographically isolated from extended kin. Geographic distance can affect how individuals relate to family members including how often they communicate. Because kinship networks provide social support related to favorable outcomes such as psychological functioning, physical well-being, and social adjustment (Bost, Cox, Burchinal, & Payne, 2002), it is important to understand if and how these relationships are maintained when family members no longer live in close proximity. This study examines how computer-mediated communication (i.e., email) may help individuals maintain family relations.

Regardless of distance, the extended family is an important resource (Johnson, 2000). Family systems theory contends that the entire family is affected by events experienced by its individual members. For that reason, connections between an individual family member and extended kin could be a positive resource for individual well-being as well as overall family functioning. Having more resources means the family is more likely to effectively manage the stressors in their lives.

Over the last decade, email has become a popular form of communication and could serve as a family resource. According to a survey conducted by the Pew Internet and American Life Project, 78% of those who went online in a typical day in 2000 sent email (Pew Internet and American Life Project, 2000). Arguably then, the Internet and email may have become an important part of many family systems.

Although we know that people are increasingly using email as a form of communication, we don’t’ know whether it is replacing more traditional forms of communication (e.g., telephone). Additionally, what influence email has on family communication and relationships is not fully understood.

A displacement effect occurs when a new communication medium affects the use of existing media. This could mean that as more people use email to communicate with their family more, they use other communication media such as telephones and letters less. Email, then, might not increase family communication, but rather it could replace other communication media. According to Kayany and Yelsma (2000), the displacement effect occurs when a new media has exactly the same function as current media. Thus, if email were only replacing other media, we would not expect to see an increase in family connectedness. Alternatively, if email is somehow different from other media, then we might expect that family communication would increase; thus, the family system would be changed.

People commonly argue that with each new advance in communication technology, people are becoming more isolated. Others contend that while we may not have as much face-to-face communication, new communication technology helps us avoid being isolated. For instance, Katz and Aspden (1997) found that long time users of the Internet reported contacting family members more often than newer Internet users. They also found that there was no difference in the membership rates of religious, leisure, and community organizations by users and non-users of the Internet. These findings demonstrate that although individuals may be interacting differently with family and neighbors now, the amount of interaction is not necessarily decreasing. In fact, they reported that the Internet was used to facilitate face-to-face meetings with family members.

To understand how email may influence family systems, we need to understand contextual features about families using email. The knowledge gap hypothesis predicts that infusions of information into an environment will lead to knowledge gain by groups with higher socioeconomic status (SES) at rates that outpace the knowledge gain of lower SES groups, leading to a growing relative gap between them (Gaziano & O'Leary, 1998; Holbrook, 2002; Tichenor, Donahue, & Olien, 1970). As the amount of information from mass media in a social system increases, people from higher socioeconomic groups tend to acquire this information more quickly than those from lower economic segments. Thus, the gap in knowledge between the two segments tends to increase rather than decrease. Following this logic, individuals with higher socioeconomic status may learn about email and the Internet faster than individuals with lower socioeconomic status.

Previous research indicates women are more likely than men to use email to maintain relationships with kin and with people who live far away. Moveover, women’s messages to family and friends are more likely to contain personal content while men are more likely to email local friends and family to coordinate joint activities (Boneva, Kraut, & Frohlich, 2001). We also know that children spend more time online than adults (Kayany & Yelsma, 2000).

Our research uses the PEW Internet and American Life Project data to examine who uses email to establish and/or maintain kinship networks and the perceived effect, if any, that email has on family relations. We also examine to what extent email may be displacing other forms of communications.

Method

The data used in this study were obtained from the PEW Internet and American Life Project. Specifically, we used the March 2000 Survey data that included extensive tracking of Internet use, email use, family connectedness, and social capital. Overall, survey results are based on telephone interviews conducted by the Princeton Survey Research Associates on a sample of 3,533 adults, 18 years of age or older, in the Continental United States during the period of March 1 to March 31, 2000. Many questions we used from this survey were asked of the 1,690 adults who are Internet users. Each day of the month, 100 interviews were completed using random digit dialing of both listed and unlisted phone numbers from telephone exchanges in the continental United States. At least 10 attempts were made to complete an interview at every household in the sample. Calls were staggered over times of day and days of the week to maximize contacts with potential respondents. Interview refusals were re-contacted at least once to try again to complete an interview. All interviews completed on any given day were considered the final sample for that day.

Non-response in telephone interviews produces known biases in survey-derived estimates because participation tends to vary for different subgroups of the population, and these subgroups are likely to vary also on questions of substantive interest. In order to compensate for these biases, the sample data were weighted in analysis. The demographic weighting parameters were derived from a special analysis of the Census Bureau’s Current Population Survey (March 1999). This analysis produced population parameters for the demographic characteristics of adults age 18 or older, living in households that contain a telephone. These parameters were then compared with the sample characteristics to construct sample weights.

Analyses for this study are based on a series of questions about using email as a means of communication in family relationships. The focus of the analyses is on self-reported behaviors of ever sending email to family members (1=no, 2=yes) and a series of questions that asked about perceived effects of using email to communicate with family members. Participants reporting that they had sent email to family members were asked to respond not true (0) or true (1) to the following items: (a) email has brought me closer to my family, (b) it is easier for family members to say frank or unpleasant things in email than in conversations, (c) email is too impersonal for communicating with family members, (d) because of email I can keep in touch with my family without having to spend as much time talking to them, (e) I have learned more about my family since we’ve been using email, (f) email has improved relationships in my family, and (g) email has added to the stress of my family. Three of the items were reverse coded so that higher scores represented a benefit. These 7 items were aggregated to create an email effect scale that had a range of 0 to 7 with higher scores indicating greater perceived benefits of emailing family members. The coefficient alpha of .674 suggests that the questions comprising the scale are internally consistent.

Two sets of independent variables were included in the analyses. First, we looked at general background information - sex (1=male, 2=female), age (number of years), education (8 categories), Hispanic ethnicity (0=no, 1=yes), race (Asian, Black, other or mixed, and White), income (8 categories), parental status (0=no, 1=yes), household structure (1=single-adult, 2=two-adult), and social support (turn to people for help, 0=no one to 3=many people). Second, we examined questions that asked about the family member that participants emailed the most. These questions included the relation to the family member which we recoded as extended (0) or immediate (1) family, location of family member recoded as in-state (0) or out-of-state (1), usefulness of email for communicating with family (1=not at all useful to 4=very useful), how often this family member was emailed, visited, and telephoned in the last week (5=every day, 4=about once a week, 3=about once a month, 2=several times a year, 1=less often), and the reasons for emailing the family member. For this last set of questions, participants were asked if they emailed the family member (a) to pass along information, (b) about something upset about, (c) to pass along joke, (d) to pass along family news, (e) about your job, (f) to get advice, (g) pass along information, and (h) get advice (0=no, 1=yes).

Logistic regression analysis was conducted to examine the relationship between email use and the sociodemographic variables and social support. Email use was coded 0 “no” and 1 “yes.” A model chi-square statistic was used to test whether the overall model was statistically significant. Odds ratios (ExpB) and associated 95% confidence intervals were used to represent the odds change in the dependent variable for a change of one unit in the independent variable, while controlling for all other variables in the model. The Wald statistic was used to test the hypothesis that a coefficient of an independent variable was significantly different from zero. (An odds ratio equal to 1 means that there is a 50/50 chance that the event will occur with a small change in the independent variable. Therefore, the odds ratio is significant if the 95% confidence interval does not include a value of 1.) Age, education, income, and social support were coded as continuous variables and the odds ratios represented the increased or decreased odds of using email associated with each unit of increase in each of these variables while controlling for all other variables. Sex, Hispanic, race, parental status, and household structure were dummy coded with male, non-Hispanic, White, non-parents, and single-adult household as the referent groups. Thus, the odds ratios represented the increased or decreased odds of using email when compared to the referent group.

Hierarchical multiple regression analysis was used to examine the relationship between email effect and each independent variable. We used the following entry format: the control variables of sex, age, education, Hispanic ethnicity, race, income, parental status, household structure, and social support (Step 1); relation to the family member, location of family member, usefulness of email for communicating with family, how often this family member was emailed, visited, and telephoned in the last week, and the reasons for emailing the family member (Step 2). Overall model significance was tested using an F statistic and the multiple coefficient of determination (R 2) represented the percentage of the total variation in email effect explained by the set of predictors. F-changes and R-square changes are used to determine if the variables added in Step 2 explain significantly more variation in email effect than the Step 1 variables alone.

After examining the overall performance of the model, individual t statistics were used to examine the individual contribution of each independent variable. Age, education, income, social support, usefulness of email for communicating with family and how often this family member was emailed, visited, and telephoned in the last week were coded as continuous variables and the parameter estimates represented the average change in email effect associated with each unit of increase in each of these variables while controlling for all other variables. Sex, Hispanic ethnicity, race, parental status, household structure, and the reasons for emailing the family member were dummy coded with male, non-Hispanic, White, non-parents, single-adult household, and “no” responses to questions concerning the reasons for emailing family member as the referent groups. Thus, the associated parameter estimates in the models represented the difference in average level of email effect for that group relative to the intercept term.

Results

A logistic regression analysis was conducted to determine who reported sending email to family members (Table 1). The results indicated the people who were more likely to have reported using email to contact family members were female, younger, non-Hispanic, and White, with higher incomes and education, living in two-adult households, and with higher levels of social support.

Table 1

Summary of Logistic Regression Analysis for Variables Predicting Email Use

 

B

SE

Wald

Exp(B)

95% CI for EXP(B)

Lower

Upper

Sex

 .16

.06

 7.25*

1.17

1.05

1.32

Age

-.04

.00

352.51***

.96

.96

.97

Education

 .45

.02

492.17***

1.57

1.51

1.63

Hispanic

-.38

.12

 10.62**

.68

.54

.86

Black

-.70

.11

 42.35***

.50

.40

.61

Asian

-.58

.26

 4.98*

.56

.34

.93

Other Race

-.58

.18

 10.19**

.56

.39

.80

Income

 .19

.02

112.24***

1.21

1.17

1.25

Parental status

-.05

.07

 .53

.95

.84

1.08

Household structure

 .29

.07

 17.98***

1.34

1.17

1.53

Social Support

 .21

.04

22.47***

1.23

1.13

1.34

c 2

1604.26***

df

11

Note: Exp(B)=exponentiated B. Age, education, income, and social support were coded as continuous variables. Sex, Hispanic, race, parental status, household structure, and the reasons for emailing the family member were dummy coded with male, non-Hispanic, White, non-parents, and single-adult household as the referent groups.

*p < .05. **p < .01. ***p < .001.

A majority (58.9%) of those who indicated that they used email said that they now communicate more with the family member they emailed the most, 1.5% said they communicated less, and 39.7% said that email had not made a difference in the frequency of communication. Email frequency had a significant, positive correlation with getting together with (r=.13, p < .01) and speaking on the phone with (r=.47, p < .01) the most frequently emailed family member.

The mean email effect score was 3.75 (SD=1.78). Overall, the results of the linear regression analysis (Table 2) indicated that the relationship between these variables and email effect were significant, F (11, 1816)=2.07, p < .05. Less than 1% of the variance in email effect was explained, however. Sex was the only significant variable with females reporting a higher level of perceived benefit to using email to contact family members. When questions about the most frequently emailed family member were added to the analysis, there was a significant increase in the amount of variance explained, R2–change=.27, F–change (12, 1793)=56.68, p < .001. Overall, we were able to explain nearly 28% of the variance in email effect scores.

Table 2

Summary of Hierarchical Regression Analysis for Variables Predicting Email Effect

 

Model 1

Model 2

Variable

B

SE B

²

B

SE B

²

Sex

 .17

.09

 .05*

-.10

.07

-.03

Age

 .00

.00

-.02

 .00

.00

 .00

Education

-.05

.03

-.04

-.01

.03

-.01

Hispanic

 .04

.17

 .01

 .11

.15

 .02

Black

 .17

.17

 .02

 .16

.15

 .02

Asian

 .22

.36

 .01

 .15

.30

 .01

Other race

-.29

.28

-.02

-.10

.24

-.01

Parental status

-.06

.09

-.02

 .02

.08

 .01

Household structure

 .23

.10

 .06*

 .07

.09

 .02

Income

-.07

.03

-.07**

-.06

.02

-.06**

Social support

 .07

.07

 .03

 .02

.06

 .01

Family member emailed

     

-.21

.09

-.05**

Usefulness of email

     

 .74

.06

 .31***

Resident status of family member

     

-.10

.09

-.03

Visit family member frequency

     

 .01

.04

 .01

Phone family member frequency

     

-.13

.04

-.09**

Email family member frequency

     

 .16

.05

 .08**

Pass along information

     

 .29

.11

 .06*

About something upset about

     

 .56

.09

 .16***

Pass along joke

     

-.03

.10

-.01

Pass along family news

     

 .43

.11

 .09***

About your job

     

 .22

.09

 .05*

Get advice

     

 .20

.08

 .06*

R2

 .01

.28

F for change in R2

2.06*

56.68***

Note: Age, education, income, social support, usefulness of email for communicating with family and how often this family member was emailed, visited, and telephoned in the last week were coded as continuous variables. Sex, Hispanic, race, parental status, household structure, and the reasons for emailing the family member were dummy coded with male, non-Hispanic, White, non-parents, single-adult household, and “no” responses to questions concerning the reasons for emailing family member as the referent groups.

*p < .05. **p < .01. ***p < .001.

Discussion

The purpose of this study was to investigate whether email, a form of computer-mediated communication, might help individuals become closer and more connected with their families. Our respondents perceived that email is a useful way to stay connected with their families and that frequent emailing is beneficial for the family system. Consistent with the knowledge gap hypothesis, individuals with more education were more likely to email than individuals with less education. In addition, younger people were more likely to email than older individuals, and females were more likely than males to say that they email family members. Boneva et al. (2001) also found that women were more likely than men to use email to maintain relationships with kin and with people who live far away.

In relation to race and ethnicity, our findings showed that Non-Hispanics were more likely than Hispanics to email, while Caucasians were more likely then African Americans, Asians, or other races to email. These findings were independent of income, a factor that is often used to explain differences between racial/ethnic groups, suggesting that there is something particular to these cultures that makes it less likely that they will use email to communicate with their families. It may simply be that these families have less of a need to use email because they are not as geographically isolated or it may be that the use of technology to communicate with family carries more of stigma among these groups.

Similarly, household structure seemed to be an important factor in determining who emails family members, regardless of SES. Individuals in two-adult households were more likely to report emailing family members than individuals in single adult households. One would think an adult who is living in a single-adults household would have more of a desire to communicate with family than an individual living with two adults. However, because they are the only adult in the household they may have more responsibilities and so they do not have the time to be sending and replying to emails. This explanation seems likely given that a measure of social support was included in these analyses.

Looking at social support, the more that an individual feels that they can turn to people for help the more likely they will be to email. Although this finding is significant, it is important to remember that causation cannot be inferred. Future research needs to identify whether individuals feel like they have more support because they emailed others and have created a support network or because they had a good support network so they emailed more often.

For those who do use email, it appears that it does more than just replace other forms of communication. Instead, email is being used in addition to other forms of communication. This suggests that email is increasing overall communication between family members. However, we did find a negative relationship between email effect and phone use showing that the more one uses the phone to contact family members, the less he/she is to believe that emailing is beneficial.

Of those who emailed family members, females were more likely to say that they felt more benefits from emailing their families than males. However, when we looked at questions relating to the primary contact person (i.e., Step 2 variables) there was no longer a difference between females and males. We did find that those who email extended family are more likely to perceive that email is beneficial to their family relationships. That is, respondents perceived that email was a useful way to communicate with family and that frequently emailing a family member is beneficial to family relationships. The content of email communication increased the perception of benefits as well. If individuals email their family, concerning something they are upset about, family news, their job, or for advice or to pass along information they were more likely to feel that they benefited by emailing.

A post hoc analysis showed that email usefulness accounted for a quarter (7%) of the variation explained in email effect scores. It is difficult to say what is cause and what is effect, however. Although a perception of email usefulness may proceed the email effect, it could also be that the email effect results from people believing email is a useful way to communicate with their families. The most likely explanation is that relationship between email effect and email usefulness is reciprocal. In other words, the belief that email is useful for contacting family members and that it will have an effect on family relationships develop simultaneously.

In summary, our findings suggest that younger Caucasian females with a higher income living in a two-adult household are more likely than other individuals to email family members. Of those who emailed family members, many reported that they felt more connected to their families; the more that they used email as a means of communication the more they benefited from the connection with their families.

Feeling connected to your family can help an individual in many areas of their lives and may help in overcoming stresses. This is consistent with assumptions from family systems theory, which states that a family is made up of many interconnected family members each affecting and influencing each other. Using email to help one individual in distress can also help other members in the family, either by feeling more connected or directly resulting from the support received. Future research needs to examine the causal nature of email use within the family as it relates to social support and overall well-being. Our study shows only some individuals are receiving the benefits of email to stay connected and receive support from their families. New opportunities are needed for individuals who do not have ready access to a computer to be able to use computer-mediated communication to develop and maintain family relations.

References

Boneva, B., Kraut, R., & Frohlich, D. (2001). Using E-Mail for Personal Relationships: The Difference Gender Makes. American Behavioral Scientist, 45(3), 530.

Bost, K. K., Cox, M. J., Burchinal, M. R., & Payne, C. (2002). Structural and Supportive Changes in Couples' Family and Friendship Networks Across the Transition to Parenthood. Journal of Marriage & Family, 64(2), 517-531.

Gaziano, C., & O'Leary, J. (1998). Childbirth and Infant Development Knowledge Gaps in Interpersonal Settings. Journal of Health Communication, 3(1), 29-51.

Holbrook, T. M. (2002). Presidential Campaigns and the Knowledge Gap. Political Communication, 19(4), 437-454.

Johnson, C. L. (2000). Perspectives on American Kinship in the Later 1990s. Journal of Marriage & Family, 62(3), 623.

Katz, J. E., & Aspden, P. (1997). A nation of strangers? Communications of the ACM, 40(12), 81.

Kayany, J. M., & Yelsma, P. (2000). Displacement Effects of Online Media in the Socio-Technical Contexts of Households. Journal of Broadcasting & Electronic Media, 44(2), 215.

Pew Internet and American Life Project. (2000). Tracking online life: How women use the Internet to cultivate relationships with family and friends. Retrieved September 20, 2003, from http://www.pewinternet.org/reports/

Tichenor, P. J., Donahue, G. A., & Olien, C. N. (1970). Mass media flow and differential growth in knowledge. Public Opinion Quarterly, 34(2), 159.

U. S. Census Bureau. (2003). Current Population Survey, March 2002; Table 14: General mobility of family householders, by type of household, age of householder, and presence and age of own children under 18: March 2001-2002. Retrieved October 5, 2003, from http://www.census.gov/population/socdemo/migration/cps2002/tab14.pdf

Note: The data for this study were provided by The Pew Internet & American Life Project. The Project bears no responsibility for the interpretations presented or conclusions reached based on analysis of the data.

 

Amy Epner and Dr. Kevin H. Gross are located at East Carolina University, Greenville, NC.