University of Central Florida Undergraduate Research Journal - Positive Outcomes of Divorce: A Multi-Method Study on the Effects of Parental Divorce on Children
US tab


The purpose of this study is to explore the effects of parental divorce on young adults' ability to form and maintain romantic relationships using surveys and face-to-face interviews. Surveys were distributed through social networking sites (i.e. Facebook, Reddit, and Webcourses) and emailed to hundreds of students currently attending the University of Central Florida (UCF), as well as to their professors with the request that they distribute the surveys to their students. Included in these surveys was a section for students whose parents divorced to leave their contact information if they were willing to meet on the main campus for a face-toface interview to discuss the specifics of their personal experience with parental divorce.


The sample for the survey consists of 233 students attending UCF, including 67 men and 166 women. Their ages ranged from 18-29 years old; the average respondent was 22 years old. Of these respondents, 45.5 percent of the total sample reported having experienced at least one parental divorce (33 men and 73 women). Most were White/Caucasian (64 percent), followed by 21 percent who identified as Latin/Hispanic, 14 percent African-Americans, and 8 percent Pacific Islanders (percentage does not equal 100 given approximately 7 percent who identified as multiracial and were included here in multiple categories).

Quantitative Data

Dependent Variables

For the quantitative portion of the study, several aspects of romantic relationship formation and maintenance, including attitudes towards relationships, number of relationships and average duration, were analyzed in this study. In order to accurately measure a respondent's attitude towards romantic relationships (the first dependent variable), a numerical scale was developed consisting of 10 five-point Likert scale statements ("I am wary of getting involved with people romantically"; "At times I think I am not a good partner"; "I am confident in my ability to maintain a relationship"; "I am able to communicate effectively with my partner"; "I feel that relationships are short-lived"; "I am motivated to make my relationship work"; "I am better at being in a relationship than most other people"; "I am sometimes afraid I will cause my relationship to end"; "All in all, I am inclined to feel that I am not fit for being in a relationship"; "I take a positive attitude toward relationships"). Depending on how the statement was phrased, answer choices would be coded to ensure that the overall direction of the score was uniform; for positive statements, the items were coded (1) strongly disagree, (2) disagree, (3) neither, (4) agree, and (5) strongly agree, while negative statements were coded inversely (i.e., (1) strongly agree, (2) agree, (3) neither, (4) disagree, and (5) strongly disagree) (see Appendix). The final scale ranges from 10-50 points, with a higher score denoting a more positive overall attitude towards relationships. For the second dependent variable, in order to measure the amount of difficulty respondents from the divorced sample experienced in maintaining a romantic relationship, the survey included two questions on the number of relationships the respondent had had following the divorce and how long on average these relationships had lasted.

Independent Variables

I measured presence of parental divorce with the question: "During your childhood and adolescence, how many times did your parents divorce (if at all)?" Answers were recoded into a new variable, with the value 0 representing respondents whose parents never divorced and the value 1 representing those respondents who had experienced at least one parental divorce. This variable was also analyzed as a continuous variable, with the values 0-4 (0 representing respondents whose parents never divorced and values 1-4 for the number of parental divorces respondents experienced respectively) in separate analyses to determine whether multiple divorces impacted outcomes. Gender was coded 0 if male and 1 if female. Respondents' age at the time of their parent's divorce was coded with the following values: 1 representing the 0-2 years old response, 2 representing the 3-5 years old response, 3 representing the 6-8 years old response, 4 representing the 9-11 years old response, 5 representing the 12-14 years old response, 6 representing the 15-17 years old response, 7 representing the 18 years or older response, and 9 representing the missing data.

Qualitative Data

The use of qualitative methodology is necessary to understand the lived experiences of children of divorce as well as to uncover possible new outcomes that cannot be measured by fixed quantitative processes. Ten respondents from the Divorced sample agreed to participate in faceto-face interviews. The interviews were semi-structured, consisting of 7 core questions and allowing respondents to delve into any aspects of their experience with which they were most comfortable. Respondents were informed prior to the beginning of the session that their contact information will remain confidential and that the interview would be audio-recorded and transcribed verbatim for quality data analysis purposes. For purposes of confidentiality, gender-neutral pseudonyms are used, as are terms such as "partner." Further, gendered pronouns do not necessarily denote the respondent's gender.

The majority of these participants experienced one parental divorce; only one experienced a second. Three participants reported that their parents had gone through multiple divorces (one in particular, Jordan, reported that her mother had been divorced twice and her father five times). The majority (n=6) of these participants reported that their fathers remarried following their most recent divorce while only one reported the mother remarrying. Eight participants reported conflict between their parents during or after the divorce and the remaining two reported no conflict.

Analytic Strategies

In order to test whether respondents from intact families have more positive outcomes with relationship formation than those from divorced families, I ran an independent samples t-test on the "romantic relationship formation difficulty" scale, comparing those whose parents had divorced with those whose parents never divorced. To test whether gender affects relationship formation, I ran independent samples t-tests, using gender as the independent variable and "number of intimate relationships (i.e. boyfriends/girlfriends)" and "length of intimate relationships" as the dependent variables for respondents who experienced at least one parental divorce. To determine whether age at time of parental divorce affects relationships, I ran an analysis of variance, using age at time of parental divorce as the independent variable and "number of intimate relationships" and "length of relationships" as the dependent variables for respondents who experienced at least one parental divorce. To determine whether higher frequencies of parental divorces has a more negative effect on relationship formation than lower frequencies, I ran another analysis of variance, using question: "During your childhood and adolescence, how many times did your parents divorce (if at all)?" and the 10 Likert scale (attitude towards relationships), using only the respondents who answered to having experienced at least one parental divorce.

For the qualitative analysis, I coded for common themes in the responses, such as family structure and duration of relationships. Participants were then sorted by these categories and compared to reveal trends and similarities between individual cases.

Findings >>