Volume 15, No. 5 Submitted: March 10, 2010

Accepted: April 27, 2010 Published: May 22, 2010


Kathleen Fuegen

Northern Kentucky University Nicole F. Endicott

Ohio State University


We tested the hypothesis, derived from the shifting standards model of stereotyping, that parenthood would polarize judgments of men’s and women’s job-related ability. One hundred thirty-five attorneys evaluated the résumé of a recent law school graduate. The résumé depicted the graduate as male or female and as either single or married with two young children. We found that a mother was held to a stricter standard for hiring than either a father or a woman without children. Results suggest earlier research conducted with undergraduates generalizes to professionals (Correll, Benard, & Paik, 2007; Fuegen, Biernat, Haines, & Deaux, 2004).


Women comprise 49% of the recipients of law degrees (NCES, 2005) but only 17% of partners in law firms (Samborn, 2005). Gender stereotyping and discrimination account for at least some of this gap (Agars, 2004; Eagly, 2005; Lyness & Heilman, 2006; Martell, Lane, & Emrich, 1996; Powell, Butterfield, & Parent, 2002; Schein, 2001). Women who become mothers face negative stereotypes about their workplace competence and commitment (Crosby, Williams, & Biernat, 2004; Hebl, King, & Glick, 2007; King, 2008). The purpose of this research is to examine how gender stereotypes affect standards for inferring job-related competence and hiring and promotion decisions for mothers and fathers. Russo (1976) and Williams (2001) have argued that there exists a cultural stereotype that a «good» mother bears primary responsibility for the care of children. A mother who works full-time is seen as deviating from gender roles, though a father is not (Etaugh & Folger, 1998). Women receive more criticism than men for spending too little time at home (and too much at work). In contrast, men receive more criticism than women for spending too much time at home (and too little at work) (Deutsch & Saxon, 1998).

A growing body of literature suggests that mothers who violate gender roles by seeking full-time employment are negatively stereotyped and discriminated against. Cuddy, Fiske, and Glick (2004) asked undergraduate participants to read vignettes describing a consultant who was either female or male and either a parent or not a parent. Participants requested and recommended the consultant less when she was a mother than a woman without children, though fatherhood did not affect a man’s chances of being requested and recommended. Consistent with predictions derived from the stereotype content model (Fiske, Cud

dy, Glick, & Xu 2002), participants rated a mother more communal (warm) but less agentic (competent) than a woman without children. Heilman and Okimoto (2008) asked undergraduate (Study 1) and graduate (Study 2) participants to evaluate a job applicant depicted as male or female and with or without children. Similar to Cuddy et al. (2004), participants recommended a mother less often than a woman without children. Fatherhood had no effect on a man’s chances of recommendation. Consistent with predictions derived from the lack of fit model (Heilman, 1983), lower ratings on agentic traits predicted lower expectations of a mother’s competence. These studies suggest that women are uniquely disadvantaged in terms of perceived job-related competence when they become mothers.

Higher Standards for Mothers than Fathers

We argue that negative outcomes for mothers reflect the setting of more stringent standards for inferring job-related competence in mothers relative to fathers and women and men without children. Testing predictions derived from the shifting standards model of stereotyping (Biernat, 2003), Fuegen, Biernat, Haines, and Deaux (2004) asked undergraduate participants to judge the level of ability (e.g., test scores and rankings on letters of recommendation) they would require of an applicant for an attorney position in order to hire him or her. Participants held a mother to higher standards than a father, and a father was held to lower standards than a man without children. Furthermore, a mother was somewhat less likely to be hired and significantly less likely o be promoted than a woman without children. Fatherhood did not affect a man’s chances of being hired or promoted, consistent with Cuddy et al. (2004) and Heilman and Okimoto (2008).

The premise of the shifting standards model is that stereotypes activate standards according to which individual members of stereotyped groups are judged. To the extent that motherhood highlights gender stereotypes (i.e., that women are warm, gentle, and caring; Spence & Buckner, 2000), mothers may be judged according to lower job-related competence standards for women in general, relative to fathers and women without children (Bridges, Etaugh, & Barnes-Farrell, 2002). Precisely because of lower expectations of competence, a mother must demonstrate more skill than a father or a woman without children to be judged equally competent.

Generalizing to a Non-student Population

One may question whether students’ judgments are similar to professionals’ judgments of job-related ability. Students are unlikely to have experience evaluating job applicants. Professionals who are parents may be more likely to hold egalitarian gender beliefs (Etaugh & Moss, 2001).

Two studies have assessed perceptions of parent and non-parent job applicants among non-student professionals. Firth (1982) mailed application letters to accounting firms in which the applicant’s gender and parental status were manipulated. Motherhood decreased the likelihood that a female applicant was contacted, but fatherhood had no effect on a male applicant’s likelihood of being contacted. Correll, Benard, and Paik (2007) mailed same-sex pairs of résumés to employers advertising marketing positions. A mother received half as many callbacks as a woman without children. Though these results suggest that professionals may be less likely to interview mothers than women without children, the results provide no information about judgment standards or hiring decisions. To fill this gap, we conducted an experiment in which we manipulated the gender and parental status of a job applicant and assessed standards for hiring, hiring decisions, and promotion recommendations among a professional sample.


We predicted that parenthood would polarize judgments of men’s and women’s job-related ability such that mothers would be held to higher standards than fathers. We base this prediction on the assumption that motherhood makes salient stereotypes that suggest that women are warm, gentle, and caring— attributes not thought to facilitate success in the workplace. To the extent motherhood makes gender stereotypes and the cultural role of care-giver salient, a female job applicant with children will be judged according to a stricter standard than a female job applicant without children. To the extent fatherhood makes the cultural role of breadwinner salient and suggests maturity, responsibility, or leadership, a male applicant with children will be judged according to a more lenient standard than a male without children.

Our predictions are unique in suggesting that men may benefit in terms of perceived job-related ability when they become fathers. Predictions derived from the stereotype content model (Cuddy et al., 2004), the lack of fit model (Heilman, 1983; Heilman & Okimoto, 2008), and expectation states theory (Correll et al., 2007; Ridgeway & Correll, 2004) suggest that parental status has no effect of judgments of men’s job-related ability. Rather, mothers are uniquely disadvantaged in the workplace because they are perceived as warm but not competent (Cuddy et al., 2004), notagentic (Heilman & Okimoto, 2008), or because motherhood is a status characteristic implying reduced performance capacity (Correll et al., 2007). We suggest that fathers are uniquely advantaged in the workplace because they are perceived as both competent and warm (Cuddy et al., 2004; Fuegen et al., 2004) and because paid employment is consistent with gender roles (Etaugh & Folger, 1998). Indeed, research has shown that fathers are held to lower standards than even the

«ideal» worker (Fuegen et al., 2004), fathers are offered a higher starting salary than men without children (Correll et al., 2007), and fathers’ work hours and wages increase when mothers temporarily leave the workforce (Lundberg & Rose, 2000).


Participants and Procedure

Participants were 135 law school graduates (74 women, 61 men). We randomly selected equal numbers of males and females from an alumni directory. Of the 588 participants contacted, 36 had undeliverable addresses, resulting in 552 potential participants. Eighty females and 68 males responded (response rate = 27%). Due to clerical error, we had to omit the responses of seven males and six females. Our sample had considerable professional experience, though males had more experience than females: males reported having an average of 21 years of experience (SD = 14.47), compared to 9 years (SD = 7.47) for females, t(83.14) = 5.55, p < .0001.

We mailed participants a recruitment script with a dollar bill attached, one of four résumés, a questionnaire, and a self-addressed stamped envelope. In the recruitment script, we stated that we were interested in how experienced decision makers evaluate job applicants. We assured participants that participation in the study was voluntary and that their answers were anonymous. We asked participants to review the résumé, complete the questionnaire, and return the completed questionnaire in the self-addressed stamped envelope. The dollar was theirs to keep regardless of whether they chose to participate.

The résumé was that of an actual law school graduate with some relevant work experience. We manipulated applicant gender via the name on the résumé («Kenneth» or «Katherine»). We chose these names because they suggest roughly equal intellectual competence and a nonspecific age (see Biernat & Fuegen, 2001). We manipulated parental status under the résumé heading

«Personal information.» We indicated the applicant was either single with no children or married with two young children (ages 5 and 3).

Dependent Measures

On the questionnaire, participants indicated what score in percentile ranking the applicant would need on a standardized ability test, letters of reference, and the Law School Admissions Test (LSAT) as well as in what percentage of his or her law school class the applicant would need to be ranked (e.g., the top 10%) to be hired as a first-year associate. These four items were standardized and combined to form the standards index (alpha = .75). Participants also indicated whether they would hire the applicant and whether the applicant would be a good candidate for promotion.



An Applicant Gender X Parental Status ANOVA on the standards index revealed only the predicted interaction, F(1, 97) = 8.75, p < .01, partial eta squared = .08 (see Table 1). A female applicant was held to higher standards when she was a parent than when not a parent, p < .02.

Standards were also higher for a mother than a father, p < .01. A father tended to be held to lower standards than a man without children, though this simple comparison was not significant, p = .10

Table 1. Standards for Hiring as a Function of Applicant Gender and Parental Status Female parent Male parent Female non-parent Male non-parent N 25 22 31 25

Mean standard for hiring 0.35 -0.27 -0.16 0.09

Standard deviation 0.37 0.77 0.96 0.64

Note. Means reflect standardized values; higher numbers indicate the setting of a stricter standard.

Hiring and Promotion Decisions

Hiring and promotion decisions were recoded as no = 0 and yes = 1. A logistic regression on hiring revealed no significant effects. Thus, there was no evidence that a mother was hired less frequently than a father or a woman without children. Regarding promotion, there was only a main effect of Parental Status, chi-square (1, N = 84) = 4.42, p < .04. Unexpectedly, a parent was more likely to be recommended for promotion than a non-parent (Ms 71.43% and 48.98%, respectively). The odds that a parent would be recommended for promotion were 2.66 times greater than the odds that a nonparent would be recommended for promotion, Wald (1, N = 84) = 4.22, p < .04 (beta = 0.98; 95% confidence interval for odds ratio = 1.05, 6.79). This effect did not depend on Applicant Gender, p = .57.


To our knowledge, this is the first study to vary the gender and parental status of a job applicant and to assess standards, hiring, and promotion decisions among professionals. Consistent with prior research, we found that mothers are negatively stereotyped in terms of job-related competence (Correll et al., 2007; Cuddy et al., 2004; Etaugh & Kasley, 1981; Firth, 1982; Fuegen et al., 2004; Heilman & Okimoto, 2008). We hypothesized that, to the extent motherhood highlights gender stereotypes, a mother will be held to higher standards than a female non-parent applicant. We found that a mother was held to higher standards than a woman without children and higher standards than a father. These findings support the shifting standards model of stereotyping. This model is unique in suggesting that because fathers are presumed to exhibit maturity, responsibility, or leadership (characteristics thought to facilitate success in the workplace), a father will be held to lower standards than a male non-parent. Also consistent with shifting standards, we found that a father tended to be held to (non-significantly) lower standards than a man without children. In contrast to earlier research showing that parental status had no effect on judgments of men’s job-related competence (Correll et al., 2007; Cuddy et al., 2004; Heilman & Okimoto, 2008), we found that parenthood benefited men (see also Haines & Bragger, 2007).

We anticipated that being held to high (or low) standards would result in a lower (or higher) likelihood of being hired. If this pattern had borne out, a mother would have been unlikely to be hired, and a father would have been likely to be hired. Instead, we found that parenthood did not harm a woman’s chances of being hired, nor did it help a man’s chances of being hired. These null effects for hiring suggest that standards do not predict decisions in a straightforward way, at least among a professional sample. Whereas Fuegen et al. (2004) found that undergraduate participants were marginally less likely to hire a mother than a woman without children, we did not find that effect in this sample. This difference in results may reflect reluctance among professionals to make decisions based on limited information. Indeed, many participants spontaneously commented that they would need more information to make a hiring decision. The same was true regarding promotion decisions: several participants reported that they would need more information to make a recommendation regarding promotion. Thus, although parents were judged as especially good candidates for promotion, caution should be exercised in interpreting this finding. The analyses for promotion are based on a smaller sample size (N = 84) than the analyses for standards (N = 101).

Future Directions

A topic in need of investigation is whether decision-makers who are parents are any more or less likely to hire a job applicant who is also a parent, relative to non-parent decision-makers. Future research examining how attitudes about work/family conflict affect standards and hiring among decision-makers with and without children is needed. Also, the ages of children may affect perceptions of job-related ability. The association between women and care-giving may be stronger for mothers of infants than mothers of school-aged children. Third, the sex-typing of the job may affect perceptions of ability. A job emphasizing communal traits (e.g., schoolteacher) may be less subject to maternal bias than a job emphasizing agentic traits.


This research contributes to a growing body of literature on the barriers mothers face as they enter or attempt to advance in the workplace. This research also highlights the value of testing hypotheses in a non-college population and increases the external validity of earlier research findings. We show that negative stereotypes of mothers’ job-related ability are no less likely to affect judgment standards among experienced than naïve observers.


Agars, M. D. (2004). Reconsidering the impact of gender stereotypes on the advancement of women in organizations. Psychology of Women Quarterly, 28, 103-111.

Biernat, M. (2003). Toward a broader view of social stereotyping. American Psychologist, 58, 1019-1027.

Biernat, M., & Fuegen, K. (2001). Shifting standards and the evaluation of competence:

Complexity in gender-based judgment and decision making. Journal of Social Issues, 57, 707-724.

Bridges, J. S., Etaugh, C., & Barnes-Farrell, J. (2002). Trait judgments of stay-at-home and employed parents: A function of social role and/or shifting standards? Psychology of Women Quarterly, 26, 140-150.

Correll, S. J., Benard, S., & Paik, I. (2007). Getting a job: Is there a motherhood penalty? American Journal of Sociology, 112, 1297–1338.

Crosby, F. J., Williams, J. C., & Biernat, M. (2004). The maternal wall.

Journal of Social Issues, 60, 675-682.

Cuddy, A. J. C., Fiske, S. T., & Glick, P. (2004). When professionals become mothers, warmth doesn’t cut the ice. Journal of Social Issues, 60, 701-718.

Deutsch, F. M., & Saxon, S. E. (1998). The double standard of praise and criticism for mothers and fathers. Psychology of Women Quarterly, 22, 665-683.

Eagly, A. H. (2005). Achieving relational authenticity in leadership: Does gender matter? Leadership Quarterly, 16, 459-474.

Etaugh, C., & Folger, D. (1998). Perceptions of parents whose work and parenting behaviors deviate from role expectations. Sex Roles, 39, 215-223.

Etaugh, C., & Kasley, H. C. (1981). Evaluating competence: Effects of sex, marital status, and parental status. Psychology of Women Quarterly, 6, 196-203.

Etaugh, C., & Moss, C. (2001). Attitudes of employed women toward parents who choose full-time or part-time employment following their child’s birth. Sex Roles, 44, 611-619.

Firth, M. (1982). Sex discrimination in job opportunities for women. Sex Roles, 8, 891-901.

Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878-902.

Fuegen, K., Biernat, M., Haines, E., & Deaux, K. (2004). Mothers and fathers in the workplace: How gender and parental status influence judgments of job-related competence. Journal of Social Issues, 60, 737-754.

Haines, E. L., & Bragger, J. (2007, May). Parental biases and penalty and leniency. Paper presented at the Caregiver Bias Working Group, Center for WorkLife Law, San Francisco, CA.

Hebl, M. R., King, E. B., & Glick, P. (2007). Hostile and benevolent reactions toward pregnant women: Complementary interpersonal punishments and rewards that maintain traditional roles. Journal of Applied Psychology, 92, 1499-1511.

Heilman, M. E. (1983). Sex bias in work settings: The Lack of Fit model.

Research in Organizational Behavior, 5, 269-298.

Heilman, M. E., & Okimoto, T. G. (2008). Motherhood: A potential source of bias in employment decisions. Journal of Applied Psychology, 93, 189-198.

King, E. B. (2008). The effect of bias on the advancement of working mothers: Disentangling legitimate concerns from inaccurate stereotypes as predictors of advancement in academe. Human Relations, 61, 1677-1711.

Lundberg, S., & Rose, E. (2000). Parenthood and the earnings of married men and women. Labour Economics, 7, 689-710.

Lyness, K. S., & Heilman, M. E. (2006). When fit is fundamental: Performance evaluations and promotions of upper-level female and male managers. Journal of Applied Psychology, 91, 777-785.

Martell, R. F., Lane, D. M., & Emrich, C. (1996). Male-female differences: A computer 58. National Center for Educational Statistics [NCES]. (2005). Digest of educational statistics. Retrieved February 22, 2007, from .

Powell, G. N., Butterfield, D. A., & Parent, J. D. (2002). Gender and managerial stereotypes: Have the times changed? Journal of Management, 28, 177-193.

Ridgeway, C. L., & Correll, S. J. (2004). Motherhood as a status characteristic. Journal of Social Issues, 60, 683-700.

Russo, N. F. (1976). The motherhood mandate. Journal of Social Issues, 32, 143-153.

Samborn, H. V. (2005). 10 years and counting: Gender disparities in salary and advancement continue in the legal workplace. Retrieved February 22, 2007, from

Schein, V. E. (2001). A global look at psychological barriers to women’s progress in management. Journal of Social Issues, 57, 675-688.

Spence, J. T., & Buckner, C. E. (2000). Instrumental and expressive traits, trait stereotypes, and sexist attitudes: What do they signify? Psychology of Women Quarterly, 24, 44-62.

Williams, J. C. (2001). Unbending gender: Why work and family conflict and what to do about it. Oxford: Oxford University Press.


This research was supported by a Coca-Cola Critical Difference for Women Faculty Grant awarded to the first author. We thank Elizabeth Haines, Jessi Smith, and Monica Biernat for their helpful comments on an earlier draft.


Kathleen Fuegen is an Assistant Professor of Psychology at Northern Kentucky University. She conducts research on how stereotypes affect selection and performance appraisals. E-mail is:

Nicole Endicott is a graduate of The Ohio State University. Her research interests are in stereotyping and marriage expectations. E-mail is:

Submitted: September 16, 2009 Revisions: March 23, 2010 Accepted: June 8, 2010


Kenneth M. Cramer Justin T. Gates Univeristy of Windsor


To examine participants ratings of perceived control, a partial-replication was conducted with the addition of culture-based covariates (horizontal/vertical, individualism/collectivism). Participants believed they would perform an undesirable task for either a short (2 minutes) or long (20 minutes) time period. Proofreading length was determined by the contents of an envelope selected either by the researcher’s coin flip (no choice) or directly by the participant (choice). Participants were told the envelopes contained either the same time periods (no control), or different time periods (control). Results indicated an illusory control effect with the provision of choice, with no impact from the cultural covariates. Implications are discussed.


It is well documented that people strive for control and mastery of their environment, and the benefits of control have received much attention (Seligman, 1975; Schulz, 1976). Much of the research in the area of control has focused on prediction, and maintained that control and prediction are nested together (Geer & Maisel, 1972; Burger & Arkin, 1980; Wortman, 1975). In other words, if one feels control over an important outcome, it also means that outcome is predictable (Schulz, 1976) — that to change the occurrence or duration of electric shocks is to also anticipate when they will occur. The resulting confound implies that one could not know whether higher perceived control was due to controllability of the outcome, predictability of the outcome, or some combination of the two.

Unconfounding Prediction and Control

Nickels et al. (1992) attempted to unconfound prediction and control, but had to reconceptualize the two. Under the reconceptualization, «prediction refers to knowing which outcome will likely occur before it occurs; control refers to exerting an influence over which outcome will likely occur» (p. 160). Under the reconceptualization, one could present controllable outcomes without prediction as «blind responses which make a difference in outcomes» (p.160). Across two experiments, the separation of prediction and control was successfully demonstrated. Participants in both prediction/control and no-prediction/control conditions provided higher scores on control-associated measures (control, influence, and responsibility) and lower helplessness ratings compared to participants in no-control conditions. The results supported the reconceptualization and the theoretical separation of control and prediction.

Control/Choice Confound

Although the aforementioned study rectified the control and prediction confound, a new confound arose. Specifically, the manipulation of control was confounded with choice as participants given controllability over the outcome also made a choice between options. This implies that one could not know whether higher perceived control was due to controllability of the outcome, choice between options, or some combination of the two (Langlois et al. 2002). Langer’s (1975) seminal work on the illusion of control demonstrated that the provision of choice alone was sufficient to produce control-related feelings. Therefore, when people are given a choice, even between options that make no difference (e.g., selecting your own lottery number before the draw), people can still report enhanced perceived control.

In an effort to disentangle controllability from choice, Langlois et al. (2002) hypothesized that participants with controllability will report higher controlrelated feelings than participants without controllability — even without being able to predict the outcome. The presence of an illusory control effect would be evident however if participants with choice (regardless of controllability) will report higher control-related feelings than participants without choice, regardless of prediction. In a two-experiment study, participants were told they would proofread medical papers for either two minutes or 20 minutes, a time period determined based on the contents of one of two envelopes (marked ‘left’ and ‘right’). Although no proofreading was actually done, participants believed they would proofread for either a more (2-minutes) or less (20-minutes) desirable time period. The envelopes were selected either by the participant (choice) or by the experimenter’s coin-flip (no-choice). The independent variables were choice and control, so that participants were randomly assigned to one of three levels therein: choice/control, choice/no-control, and no-choice/no-control. In these conditions, choice-participants selected between envelopes, wereas no-choiceparticipants received the envelope determined by the experimenter’s coin flip. Conditions with control involved different outcomes (e.g., different time periods), whereas conditions with no control involved the same outcomes (i.e., identical time periods in the envelopes).

The dependent variables for this study were participants’ self-reported feelings of prediction, control, responsibility, and helplessness. The dependent measures questionnaire was completed after the experimental manipulation. In order to hold prediction of the outcome constant at no-prediction, knowledge of the outcome was withheld from participants until they had completed the questionnaire. Results indicated that to render higher feelings of perceived control, one’s choice must make a difference in the outcome (i.e., outcomes must be different). Specifically, participants in the control/choice condition felt more control and responsibility than the other two conditions, which did not differ significantly from each other. Moreover, participants with control (different outcomes) felt more influence than participants without control (same outcomes). Helplessness did not vary significantly among the groups. Overall, the illusory control hypothesis of elevated feelings of perceived control with simply the provision of a choice (where control and choice were no longer confounded) was disconfirmed.

Control across Cultural Dimensions

The present study is a replication of Langlois et al. (2002) with the consideration of cultural variables, namely the Western-based individualist (personal achievement) vs. Eastern-based collectivist (group accomplishment) dimension. Individualism involves a self-concept that is both autonomous and unbound by one’s in-group (e.g., close family and friends). The individualistic person’s goals do not necessarily overlap with the group’s goals and relationships are only maintained if the benefits and costs are balanced. A collectivistic orientation involves a self-concept that is bounded to the group. Goals typically overlap with the group and when a discrepancy exists, group goals are a priority. Relationships are of the utmost importance and are maintained at all costs (Singelis, Triandis, Bhawuk, & Gelfand, 1995; Triandis, 1995).

Triandis, McCusker, and Hui (1990) described attributes they called pure individualistic and collectivistic types. Collectivistic people pay more attention to their in-group and behave differently in regards to that group compared to those who are individualistic. The most important in-group is the family. For those who are individualistic, the in-group and the out-group are less defined and therefore, they do not behave as differently between these groups. As previously described, when personal goals are discrepant with group goals, collectivistic people place emphasis on group goals whereas individualistic people place emphasis on personal goals.

Given the importance of the in-group for those who are collectivistic, norms are largely derived from one’s in-group. For individualistic cultures, personal likes and dislikes are more important.

Triandis et al. infer that group norms largely determine social behaviour in collectivistic cultures whereas attitudes are more important in individualistic cultures.

Indeed, differences in control related feelings between collectivistic and individualistic participants seem plausible. Ji, Peng, and Nisbett (2000, p. 944) state that Westerners place so much importance on control «they often fail to distinguish between objectively controllable and uncontrollable events, tend to perceive more control than they actually have, and report mistakenly high levels of predictability of events.» Furthermore, Iyengar and Lepper (1999) demonstrated that American children were more motivated to perform a task when they chose that task themselves. In contrast, Asian American children were more motivated to solve a task that their mother chose. In a study on decision-making, Kim and Drolet (2003) investigated the effects of cultural assumptions of choice and uniqueness on the tendency to seek variety in choices. The results showed cultural differences in the likelihood of variety-seeking in regards to choice rules. Those from individualistic cultures demonstrated a tendency to vary choice rules whereas those from collectivistic cultures did not. In short, people from different cultures may respond differently when given control over a choice.

However, Triandis and Gefland (1998) further distinguish the individualistcollectivist dimension according to both horizontal and vertical subcategories. In general, horizontal cultures value equality among members, status is fairly even, and members are seen as similar to one another. Vertical cultures demonstrate inequality and status hierarchy; members are seen as different from one another. This creates four possible types based on the two cultural dimensions. In horizontal collectivist cultures, the self is viewed as merged with the in-group; members of the in-group are viewed as similar, and equality is emphasized. For vertical collectivist cultures, the self is viewed as an aspect of the in-group, where members are seen as different from one another; inequality is both accepted and expected. For horizontal individualist cultures, the self is viewed as autonomous but equal to other members of the group. Finally for vertical individualist cultures, members are viewed as different, status inequality is expected, and competition remains important (Singelis, et al., 1995).

In the present study, whereas no specific hypotheses were made with respect to culture, it was expected that each of the four cultural covariates would influence the extent to which the three from one another. That is, after accounting for the cultural covariates of horizontal/vertical and individualist/ collectivist, that the mean differences — expected to be highest among participants with both control and choice — would alter significantly.



Eighty-seven University of Windsor undergraduate students (65 women and 22 men) volunteered to participate in order to attain partial course credit. The mean age was 21.54 years.

The sample consisted of 34 Caucasians, 21 Asians, 13 Middle Easterners, 9 African Americans, and 10 participants of mixed or other ancestry. All participants were Canadian-born.


Two white envelopes with «LEFT» printed on one and «RIGHT» printed on the other contained index cards with the time periods «2 minutes» or «20 minutes» printed on the back in light pencil so as not to be revealed through the envelope. Medium-sized signs (8.5 x 11 inch) were used to provide controland choice-relevant information to participants. Signs indicating control-relevant information read either «THE ENVELOPES CONTAIN THE SAME TIME PERIOD» or «THE ENVELOPES CONTAIN DIFFERENT TIME PERIODS.»

Signs indicating choice-relevant information read either «YOUR TIME PERIOD WILL BE DETERMINED BY THE ENVELOPE YOU SELECT» or «YOUR TIME PERIOD WILL BE DETERMINED BY THE THE ENVELOPE SELECTED FROM AN EXPERIMENTER-FLIPPED COIN.» These grouprelevant signs were placed in the room with the participant, and the experimenter ensured that participants could read and understand them.

A two-dollar Canadian coin — with the words «LEFT» and «RIGHT» affixed to each side on white paper — was used. A sheet of paper contained a list of medical verbiage under the heading «Commonly Misspelled Medical Terms.» Under the list was an excerpt from a medical paper.

Two large medical textbooks with clearly visible titles were placed on the table in front of the participants.

Design and Procedure

For the present study, participants were led to believe they were participating in a study designed to investigate the effects of personality on proofreading accuracy. All participants received the same instructions from the researcher based on their experimental condition. Upon arrival, participants completed a consent form, and were advised they would proofread medical papers for either a short (2-minute) or long (20-minute) time period. They were then presented with a list of fictitious medical terms and several sentences from a sample medical paper. Prior to the experimental manipulation, participants completed a horizontal/vertical and individualism/collectivism measure (Triandis & Gelfund, 1998). This 32-item questionnaire asked participants to rate their level of agreement or disagreement with several statements on a scale from one to nine (1-strongly agree, 9-strongly disagree). Triandis and Gelfund (1998) report adequate internal consistency and supportive validity for these constructs. Addition of these covariates represents an extension of the replicated study.

Participants were tested individually in a private room. An experimental design with random assignment to three conditions was used. Each condition involved selecting one of two envelopes, and participants were told that they would proofread medical papers for the amount of time indicated in the chosen envelope (either 2 or 20 minutes). In these conditions, participants with control over the outcome were presented with envelopes containing different time periods (outcome); participants with no control were presented with envelopes containing identical time periods (outcomes). Participants in choice conditions selected an envelope themselves whereas participants in no-choice conditions had the envelope selected for them by the experimenter’s coin-flip. Thus, participants were randomly divided into three independent groups: Participants in the control/choice condition (n = 31) chose between two envelopes which contained 2 and 20 minute time cards and they did not know which envelope contained which time period. In the no-control/choice condition (n = 27), participants chose between the envelopes and were told that both contained the same amount of time (either 2 or 20 minutes) but they did not know which.

In the no-control/no-choice condition (n = 29), participants were told both envelopes contain the same time period but selection was made by the experimenter’s coin flip. Participants were deliberately not told what time period they received in order to hold prediction constant at no-prediction.

After experimental manipulations and prior to opening the envelope, participants completed a dependent measures questionnaire. Participants rated their levels of perceived control, responsibility, influence, and helplessness on a scale from 1 to 5 (1-not at all, 5-to a great extent). This questionnaire also included demographic questions as well as manipulation checks to assess

their knowledge of their prediction, control, and choice in their given experiemental group; participants who incorrectly answered any of the main manipulation checks were disgarded from analysis. One final check assessed whether participants wanted to proofread for the short (2-minute) time period. The entire process took approximately 30 minutes. After the experiment was completed, participants were fully debriefed by the researcher and informed they would not be proofreading any papers.


A bivariate correlation indicated the dependent variables of perceived control, influence, and responsibility were all negatively correlated with helplessness and positively correlated with each other (see Table 1). An ANOVA with group (choice/control, choice/no control, no choice/no control) as the independent variable indicated nonsignificant group differences for both influence, F (2, 84) = 2.580, p = .082, and helplessness, F (2, 84) = 1.702, p = .189, but significant group differences for responsibility, F (2, 84) = 3.748, p = .028, and perceived control, F (2, 84) = 8.252, p = .004. Ryan-Einot-Gabriel-Welsch F multiple comparison tests were conducted to investigate the specific differences between the three levels of the independent variable (choice/control, choice/no control, no choice/no control) for control and responsibility. Results indicated that participants with both choice and control reported higher feelings of perceived control (M = 2.90, SD = 1.14) and responsibility (M = 3.13, SD = 1.28) than the no-control groups. In addition, participants with choice but no control reported higher feelings of both perceived control (M = 2.52, SD = 1.4) and responsibility (M = 2.74, SD = 1.48) than participants with no choice and no control (M = 1.86, SD = 1.03; M = 2.17, SD = 1.31, respectively). This would suggest the presence of an illusion of control, produced by choice without control.

In an assessment of the covariates, the bivariate correlation matrix revealed nonsignificant correlations between each of perceived control, influence, and helplessness for all four cultural covariates (horizontal/vertical and individualism/ collectivism). As shown in Table 1, there was a significant correlation between horizontal-collectivism and responsibility, r(85) = .25, p = .022; with vertical collectivism approaching significance, r(85) = .20, p = .064. Repeating the means analysis for responsibility with the covariates included, an ANCOVA indicated that after accounting for the horizontal/vertical and individualism/collectivism covariates, the results from the ANOVA were no different, F (2, 80) = 4.535, p = .014. In other words, inclusion of the cultural covariates did not alter the original findings.


The results of the present study did not fully replicate those of Langlois et al. (2002). They hypothesized that «even without prediction of outcome, participants with choice of option will report higher-control associated measures than participants with no choice of option» (p. 168).

Participants with both choice and control felt more control and responsibility than participants with choice but no control and participants with neither choice nor control. Furthermore, participants with control felt more influence than those without control. These findings disconfirmed the hypothesis that simply making a choice will result in control-related feelings.

They concluded that «to feel control, one’s actions (e.g., choice of envelope) should make a difference in one’s outcome (e.g., proofreading time)» (p. 170). That is, in order to render feelings of perceived control, simply making a choice was not sufficient; one’s choice had to make a difference in the desired outcome. Finally, the present results disconfirmed the hypothesis that inclusion of cultural dimensions of horizontal/vertical, individualism/collectivism would alter

the resulting group differences.

Although one of the cultural covariates (horizontal-collectivism) was significantly related to one of the control-associated dependent measures (responsibility), its consideration as a covariate did not significantly alter the group differences.

In the present study, although we similarly found the highest ratings of perceived control and responsibility among those with both control and choice, and the lowest ratings among those with both no-control and no-choice, we also uncovered a unique effect. That is, participants with choice but no control perceived less control and responsibility than those with control but more than those without. In short, we identified an illusory control effect wherein the presence of choice without control can significantly enhance feelings of control and responsibility among participants (Langer, 1975). Although Ji et al. (2002) would contend that Westerners are more susceptible to illusory control due to the importance placed on personal control in Western Society, past research has shown that choice alone does not necessarily produce those feelings (Cramer & Perrault, 2006; Langlois et al., 2002).

Further research is needed to investigate the effects of choice on illusory control. Specifically, conditions under which choice alone leads to illusory control can be investigated. Also, research can investigate differences in susceptibility to illusions of control based on cultural differences (for example, individualism and collectivism). A limitation exists concerning the crossing of choice with control.

Without implementing a no-choice/control condition, choice and control are not completely crossed (Langlois et al. 2002). This condition poses methodological difficulties and thus was not included in the study. Future research can investigate situations in which participants feel control without making any choices.

When considering the sample for this study, it is important to note that it was collected at a Canadian university. Although the majority of this sample was non-Caucasians, Western values and belief systems could be internalized by these participants. Perhaps a cross-cultural sample consisting of participants currently from different cultures will yield different results. Furthermore, there is a potential limitation concerning the experimental manipulation. Although it was assumed participants prefer to proofread for a short period (e.g., 2 minutes), perhaps this sample did not have a vested interest in that particular outcome.


Ken Cramer is a full professor and 3M teaching fellow at the University of Windsor in Southwestern Ontario, Canada. (there is more contact information below).

Justin Gates is a graduate psychology student at Lakehead University in Thunder Bay, Ontario, Canada.


Burger, J. M., & Arkin, R. M. (1980). Prediction, control, and learned helplessness. Journal of Personality and Social Psychology, 38, 482-491.

Cramer, K. M., Nickels, J. B., & Gural, D. M. (1997). Uncertainty of outcomes, prediction of failure, and lack of control as factors explaining perceived helplessness. Journal of Social Behavior and Personality, 12, 611-630. Cramer, K. M., & Perrault, L. A. (2006). Effects of predictability, actual controllability, and awareness of choice on perceptions of control. Current

Research in Socail Psychology, 11, 111126.

Geer, J. H., & Maisel, E. (1972). Evaluating the effects of the predictioncontrol confound. Journal of Personality and Social Psychology, 23, 314-319.

Iyengar, S. S., & Lepper, M. R. (1999). Rethinking the value of choice: A cultural perspective on intrinsic motivation. Journal of Personality and Social Psychology, 76, 349-356.

Ji, L., Peng, K., & Nisbett, R. E. (2000). Culture, control, and perception of relationships in the environment. Journal of Personality and SocialPsychology, 78, 943-955.

Kim, H. S., & Drolet, A. (2003). Choice and self-expression: A cultural analysis of varietySocial Psychology, 85, 373-382.

Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology, 32, 311-328.

Langlois, M. W., Cramer, K. M., & Mohagen, R. B. (2002). Isolating the effects of control, choice, and prediction. Current Research in Social Psychology, 7, 163-181.

Nickels, J. B., Cramer, K. M., & Gural, D. M. (1992). Toward unconfounding prediction and control: Predictionless control made possible. Canadian Journal of Behavioral Science, 24, 156170.

Paterson, R. J., & Neufeld, R. W. J. (1995). What are my options? Influence of choice availability on stress and the perception of control. Journal of Research in Personality, 29, 145167.

Presson, P. K., & Benassi, V. A. (1996). Illusion of control: A meta-analytic review. Journal of Social Behaviour and Personality, 11, 493-510.

Schulz, R. (1976). Effects of control and predictability on the physical and psychological wellbeing of the institutionalized aged. Journal of Personality and Social Psychology, 33, 565-573.

Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco: W.H. Freeman.

Singelis, T. M., Triandis, H. C., Bhawuk, D. P. S., & Gelfand, M. J. (1995). Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. Cross-Cultural Research, 29, 240-275.

Triandis, H. C. (1995). Individualism and collectivism: New directions in social psychology. Boulder, CO: Westview Press.

Triandis, H. C., & Gelfand, M. J. (1998). Converging mesurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology, 74, 118-128. Triandis, H. C., McCusker, C., & Hui, H.

C. (1990). Multimethod probes of individualism and collectivism. Journal of Personality and Social Psychology, 59, 1006-1020.

Wortman, C. B. (1975). Some determinants of perceived control. Journal of Personality and Social Psychology, 31, 282-294.

Submitted: December 11, 2009 Revision: May 13, 2010

Accepted: May 15, 2010


Nick J. Richardson & Christopher C. Barnum St. Ambrose University


This article introduces a theory describing the relationship between factors that increase social isolation and deviance. The theory is examined in the context of virtual visitation. We integrate social exchange, anomie, and strain theories to argue that as communication is impeded between two actors, the less satisfied either will be with the communication and thus, the more alienated they may become. Social alienation, in turn, may make actors more inclined to commit deviance. Findings have important implications for a broad set of applied settings including correctional facilities that have replaced in-person contact visits with virtual visits.


America’s prison and jail population is rapidly increasing. Currently, well over two million adults are incarcerated in federal, state and local correctional facilities (BJS, 2008). For nearly all these inmates, face-to-face contact with the outside world is significantly restricted. Consequently, inmates are isolated from their families and friends and must rely on periodic visits to maintain personal contact. In recent years however, a growing number of America’s correctional facilities have replaced in-person contact visits with virtual visits. For inmates in these institutions, meetings are no longer face-to face, but instead take place over a video screen.

The impact of this virtual form of visitation on the emotional attachment between inmates and their visitors has not been studied and is not currently known. It is understood however, that virtual communication lacks «media richness» in comparison to face-to-face interaction (de Pillis & Furumo, 2007). Communication using a video screen diminishes the visibility of non verbal cues including subtle facial expressions, body posture or faint olfactory signals. It also eliminates the possibility of interpersonal contact. Research has shown that the consequences associated with these deficits include misinterpreted messages, decreased trust and commitment as well as diminished satisfaction wih the interaction (Straus & McGrath, 1994; Watson-Manheim & Belanger, 2002).

In what follows, we introduce a theory describing the relationship between factors that increase social isolation and deviance. We argue that over time, any feature of a social exchange setting hat creates impediments to interaction, including aspects of communication that isolate actors by decreasing the fulfillment of the interaction between them and significant others, also diminishes the strength of social bonds between actors and their visitors. It is well understood that people who experience weakened bonds or attachments with others are more likely to engage in antisocial behavior than their counterparts (Hirschi, 1969). Consequently, impediments to interaction contribute to increases in behavior that violates accepted mores. We believe that virtual visitation is one such impediment, and use it to illustrate this process. In the following sections, we review the relevant literature and present our theory.


Exchange Theory

The genesis of social exchange theory can be traced to the works of several early contributors including, Homans (1961), Thibaut and Kelley (1959), Blau (1964), and Emerson (1962, 1972). These works blended rational choice and reinforcement processes to explain how actors in a social setting exchange valued items.

Social exchange differs from purely economic theory in a fundamental way. Traditionally, micro economic theory assumes that sequential exchanges are independent events. Social exchange theory, however, takes the recurring exchange relation as its subject matter (Molm & Cook, 1995). Exchange is a joint activity that is often built up over time. The perspective assumes that self interested actors exchange with other self interested actors in order to reach goals they could not achieve alone (Lawler & Thye, 1999). The outcomes of exchange serve as reinforcement contingencies that foster the connection. Consequently, the exchange relationship is the smallest unit of analysis.

Recently, Lawler (2001) proposed an affect theory of social exchange. The theory moves beyond current theorizing that conceptualizes rational choice and reinforcement as the primary processes that mediate structural exchange opportunities. Instead, the theory argues outcome of exchange produces global emotions which are involuntary, primitive and nonspecific feelings that are not immediately attached to particular social objects (Lawler, Thye & Yoon, 2000). If the outcome of exchange is successful, actors feel good; if the outcome is unsuccessful, they feel bad (Lawler & Thye, 1999). The emotions generated by exchange function as internal self reinforcing or punishing stimuli. Actors try to reproduce positive global feelings and avoid negative global emotions. These emotions also trigger cognitive efforts to explain their source. Actors involved in joint exchange tasks attribute these emotions to their exchange relations or groups. Lawler (2001) notes that the fundamental implication of these assumptions is that successful exchange increases group cohesion and unsuccessful exchange reduces it. Cohesion is defined as the strength and durability of person to person and person to group relations.


A long tradition of work in sociology describes and explains the association between social integration and antisocial behavior. Durkheim specifies how deviance is mediated by two social processes. Egoism is a feeling of purposelessness due to a lack of social integration and anomieis negative emotion and cognition generated by weak normative regulation of behavior. Although there is slippage related to the meanings of these terms, (see Bearman, 1991), several theorists argue that egoism and anomie are two sides of the same coin (Gibbs & Martin, 1964; Johnson, 1965). For them, social integration is a precondition for regulation. The latter cannot exist without the former. Durkheim felt that the more individuals interact in social systems subsumed by strong social relationships, the more likely they are to become attached to others and the less likely they are to carry out antisocial behavior (Stockard & O’Brien, 2002).

Although the study of homicides and suicides has developed in different subdisciplines and & O’Brien, 2002), Durkheim believed the causes of suicide and homicide in complex modern societies are «parallel» and stem from anomie or egoism (Stockard & O’Brien, 2002). Durkheim specified four sources or factors that lead to suicide. Egoistic suicide results from excessive individualism and occurs when the ties connecting the individual to society or others are weak. In essence, there are so few bonds linking the person to society that nothing is left to prevent the person from self destructive behavior. Altruistic suicide is caused by too much social integration. These people are so firmly connected to society that they are willing to sacrifice their lives for the good of the collective. Durkheim specified several categories of anomic suicide. Although somewhat varied, each of these categories share the common assumption that suicide occurs when a sense of normlessness generates a disjunction between goals and means until finally, people discover they lack the wherewithal to fulfill their needs. Fatalistic suicide results from excessive regulation of norms. Durkheim believed this was a rare phenomenon that occurred when the individual perceived no hope of relief from the oppressive regulation brought on by society. Although Durkheim’s discussion of homicides is somewhat indistinct and ambiguous, his core argument specifies that (at least for modern social systems), the more strongly individuals are integrated into and regulated by society, the less likely they will be to experience any type of lethal violence whether it be suicide or homicide (Stockard & O’Brien, 2002).

Strain Theory

This tradition traces its roots to Durkheim and holds that delinquency and deviance occur when individuals experience strain because they are unable to achieve their goals through legitimate means (Merton, 1938; Cohen, 1955; Cloward & Ohlin, 1960). Early formulations, including Merton, argued that deviance resulted from anomie or normlessness produced by a disjunction between economic goals and structurally determined channels for achieving these outcomes.

When faced with this disjunction, people may choose non-normative methods for achieving the culturally accepted goal of becoming wealthy. These «innovators» resort to crime or deviance to manage strain. More recently, Agnew (1992) extends these ideas by suggesting first, that negative experiences are key in generating deviant behavior because, among other things, strain or the failure to achieve positively valued outcomes produces negative emotion which leads to deviant behavior (Kaufman, Rebellon, Thaxton & Agnew, 2008). And second, that individuals who are subject to repetitive strain are more likely to commit deviance or crime than others (Agnew, 1992).

Social Control Theory

It is well understood that the social control theoretical tradition (Hirschi, 1969) also builds upon Durkheim’s work in explaining deviant criminal behavior (Stockard & O’Brien, 2002). Both traditions argue that weakened attachment is fundamental in generating antisocial behavior. Following Durkheim, Travis Hirschi, in his landmark work Causes of Delinquency (1969), argued that the more people are connected to society and other social institutions by social bonds, the less likely they will be to engage in criminal behavior.

According to Hirschi, four types of social bonds keep individuals from antisocial behavior. Involvement is the extent of participation in «conventional activities» such as school, work and sports. The more people are involved in these activities, the less time they have for deviant acts. Commitment is the investment of time and energy in conventional activities. An increase in commitment is inversely related to criminal or deviant behavior. Belief is the idea that the rules, laws, and norms of society should be obeyed. The strength of belief in conformity partially determines whether an individual commits a criminal act. Attachment is an enduring, strong, affective relationship between two or more people. This association is characterized by a sense of caring and respect for the opinions and feelings of the other person or group. An inverse relationship exists between attachment and criminality.

Research indicates that of the four types of bonds, attachment is the strongest predictor of criminal behavior (Bernard, Snipes, & Gerould, 2010; Agnew, 1991; Empey & Stafford, 1991; Jensen & Rojek, 1992; Junger-Tas, 1992). For example, Hirschi (1969) found that attachment, measured as communication between child and caregiver, or attachment measured as affinity to school and peers, was important for control. In family relationships distinguished by a lack of communication, children worry less about their parents’ disapproval than in more communicative family units. This indifference increases chances for deviance. Likewise, juveniles who do not like school, their teachers, or peers tend to commit more delinquent acts than other children. Simply put, this research suggests that people are less likely to commit crime if they believe it may jeopardize the affection and respect of those people in their lives who are important to them.


In this section we introduce our theory. The theory consists of three logically connected assumptions and a set of scope conditions. Each assumption is supported by relevant theory and research.

Scope Conditions:

All assumptions apply from the perspective of a focal actor designated as ‘P.’ (2) The interaction occurs in social exchange settings where (3) a focal actor perceives an opportunity to repeatedly exchange with at least one same specific other and where (4) the outcome of the exchange is uncertain. (5) Finally, the focal actor and another are exchanging tangible items such as money or nontangible, socially valued items such as friendship and esteem. Assumptions: The theory is expressed through three logically related assumptions that explain how the structural features of exchange generate emotions that lead to the development of deviance under the scope conditions.

Assumption 1: The more communication is impeded between actors ‘P’ and ‘O’, the less ‘P’ will be satisfied with the outcome of interaction with ‘O’.

An impediment to communication is anything that blocks, delays, strains, or reduces the frequency of, or interferes with, the interaction. Communication researchers have found that these factors are often associated with lower satisfaction (de Pillis & Furumo, 2007). Researchers comparing virtual to faceto-face interaction (for example in business teams or classrooms) have found that virtual communication is generally equally effective in accomplishing goals, but members of face-to-face interaction report significantly higher levels of satisfaction with the interaction (Warkentin, Sayeed, & Hightower, 1997; Arbaugh, 2000).

Assumption 2: The less ‘P’ is satisfied with the outcome of interaction with ‘O,’ the more P experiences alienation from O.

Alienation is defined as emotional detachment or withdrawing personal affection from a social relation. Support for this assumption comes principally from Lawler’s (2001) affect theory of social exchange. This approach argues that three fundamental processes occur in exchange settings: (i) the act of exchange generates emotions among participants, (ii) actors attribute these emotions to the exchange relationship itself and (iii) these emotions serve as internal reinforcement contingencies that either increase or reduce group cohesion. Unsuccessful exchange generates negative emotion and a hedonistic or self serving bias that reduces cohesion between exchange partners.

Assumption 3: the more P experiences alienation from O, the more P is inclined to commit deviance. occurs when an actor breaks an accepted normative standard of the setting. Norms include formally enacted rules such as laws and regulations and informal behavioral standards that are not codified into law. Deviance ranges from offenses against rules that are severely sanctioned, such as personal violence against others, to infractions of informal norms that are not officially punished, such as cheating during a casual game of golf.

When actors are detached or withdrawn from social relations, social control theory and research suggests that deviance is likely to occur. Weakened bonds reduce the degree that people care about the thoughts and opinions of others, which increases the probability of normative rule violations. Informal social control by friends, family and neighbors is weakened in these circumstances because the offender is less likely to experience a feeling of shame (Braithewaite, 1989). Consequently, any social characteristic that decreases interdependency among people is likely to increase the probability of deviance (Hamilton & Rauma, 1995).

The theory generates several derivations which are logical consequences of the assumptions of the theory. The following one is of particular interest. Primary Derivation: The more communication is impeded between P and O, the more P is inclined to commit deviance.

This derivation follows from assumptions 1 and 3 and stems primarily from ideas found in strain theory. This tradition holds that delinquency and deviance occurs when individuals are unable to achieve their goals through legitimate means (Merton, 1938; Cohen, 1955; Cloward & Ohlin, 1960). In this context, individuals are unable to achieve satisfaction from interaction. Agnew (1992) extends these ideas by suggesting that negative experiences generate deviant behavior because the failure to achieve positively valued outcomes produces negative emotion which leads to deviant behavior (Kaufman, Rebellon, Thaxton & Agnew, 2008).


Social exchange theory, control theory and strain theory each successfully account for responses to emotional reactions generated by social structure. However, the theories take different explanatory routes. Social exchange focuses on responses to exclusion from exchange, control theory directs attention towards emotions generated by alienation and strain theory examines how anomie affects behavior. Integrating elements of each of these theories opens an array of potential new applications without subverting the theory’s basic assumptions. Our integrated theory argues that impeded communication disrupts the exchange process, and this in turn leads to deviance that is a result of strain and alienation. Our theory allows for an array of new predictions that have important implications for applied settings, including one of the most important: an increasing number of correctional facilities that have replaced in-person contact visits with virtual visits.


Agnew, R. (1991). The interactive effects of peer variables on delinquency.

Criminology, 29(1): 47-72.

Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology. 30(1): 47-87.

Arbaugh, J. B. (2000). Virtual classroom characteristics and student satisfaction with internet-based MBA courses. Journal of Management Education, 24: 32-54.

Bearman, P. S. (1991). Desertion as localism: Army unit solidarity and group norms in the U.S. Civil War. Social Forces, 70: 321–42.

Bernard, T. J., Snipes, J. B., & Gerould, A. L. (2010). Vold’s Theoretical Criminology. New York, NY: Oxford University Press, Inc.

Blau, P. (1964). Exchange and Power in Social Life. New York: John Wiley & Sons.

Braithwaite, J. (1989). Crime, Shame and Reintegration. New York: Cambridge University Press.

Bureau of Justice Statistics. (2008). U.S. Department of Justice, Office of Justice Programs. Washington, D.C.

Cloward, R. A. and Ohlin, L. E. (1960). Delinquency and opportunity: a theory of delinquent gangs. Glencoe: Free Press.

Cohen, A. K. (1955). Delinquent Boys: The Culture of the Gang. New York: Free Press.

de Pillis, E. G., & Furumo, K. (2007). Technical opinion: Counting the cost of virtual teams. Communications of the ACM, 50:93-95.

Emerson, R. M. (1962). Power-Dependence Relations. American Sociological Review 27: 31-41.

Emerson, R. M. (1972a). Exchange theory part I: A psychological basis for social exchange. In J. Berger, Morris, Jr., & B. Anderson (Eds.), Sociological Theories in Progress (pp. 38-57) Boston, MA: Houghton Mifflin.

Emerson, R. M. (1972b). Exchange theory, part II: Exchange relations and network structures. In J. Berger, M. Zelditch, & B. Anderson (Eds.) Sociological Theories in Progress, Vol. 2. (pp. 38-87). Boston, MA: Houghton Mifflin.

Empey, L. T. & Stafford, M. C. (1991). American Delinquency: Its Meaning & Construction. Belmont, CA: Wadsworth

Gibbs, J. P. & Martin, W. T. (1964) Status Integration and Suicide. Eugene, OR: University of Oregon Books.

Hamilton, V. L., & Rauma, D. (1995). The social psychology of deviance and law. In F. Cook & House (Eds.) Sociological Perspectives on Social Psychology. (pp. 524 — 547) Boston, MA: Allyn and Bacon.

Hirschi, T. (1969). Causes of Delinquency. CA: University of California Press.

Homans, G. (1961) Social Behavior. New York: Harcourt, Brace & World.

Jensen, G. F. & Rojek, D. G. (1992). Delinquency and Youth Crime (2nd ed.). Prospect Heights, IL: Waveland Press.

Johnson, B. D. (1965). Durkheim’s one cause of suicide. American Sociological Review, 30: 875–86.

Junger-Tas, J. (1992). An empirical test of social control theory. Journal of Quantitative Criminology, 8: 9-28.

Kaufman, J. M., Rebellon, C. J., Thaxton, S., & Agnew, R. (2008). A general strain theory of racial differences in criminal offending.

Lawler, E. (2001). An affect theory of social exchange. American Journal of Sociology, 107: 321-352.

Lawler, E. & Thye, S. (1999). Bringing emotion into social exchange theory. Annual Review of Sociology, 25:217-44.

Lawler, E., Thye, S., & Jeongkoo, Y. (2000). Emotion and group cohesion in productive exchange. American Journal of Sociology, 106: 616-57.

Merton R. K. (1938). Social structure of anomie. American Sociological Review, 3: 672-682.

Molm, L. D. & Cook, K. S. (1995). Social Exchange and Exchange Networks. Pp. 209-235 in Karen S. Cook, Gary A. Fine and James S. House (eds.) Sociological Perspectives on Social Psychology. Boston: Allyn and Bacon.

Stockard, J., & O’Brien, R. M. (2002). Cohort variations and changes in agespecific suicide rates over time: Explaining variations in youth suicide. Social Forces, 81: 605-642.

Straus, S. G., & McGrath, J. E. (1994). Does the medium matter? The interaction of task type and technology on group performance and member reactions. Journal of Applied Psychology, 79: 87-97.

Thibaut, J. & Kelley, H. (1959). The Social Psychology of Groups. New York: Willey.

Warkentin, M., Sayeed, L., & Hightower, R. (1997). Virtual teams versus face to-face teams: An exploratory study of a web-based conference system. Decision Sciences, 28(4): 97.

Watson-Manheim, M. B. and Bélanger, F., (2002). Support for communication-based work processes in virtual work, E-Service Journal, 1(3): 61-82.