Ring by Spring?: Attitudes Toward Marriage Among College Students
Monika R. Hartsel 2
Jacob L. Koch 2
Carol Ann Pedersen 2
ABSTRACT
Although many studies have investigated intimate relationships on college campuses, this research looks at college students' attitudes and expectations regarding marriage. Using survey data from a random sample of students at a small liberal arts college in the upper Midwest, our research examines the degree to which students expect to find their spouse on their college campus. First, we explore how students' romantic attitudes are associated with hoping to meet their spouse at college. Secondly, we look at how speculations about the percentage of college students who marry their classmates are associated with students' desire to meet their spouse at college. We found that even though 90% of students believed that other students will marry another student from the same college, only 40% expected to meet a spouse at their college. In addition, students with more romantic attitudes were more likely to expect to meet their spouse at their college.
INTRODUCTION
St. Olaf College, a small liberal arts school in Northfield, MN, sits on a picturesque hill in Southern Minnesota. Though a magnificent sight in any season, autumn is arguably when Olaf achieves its highest aesthetic beauty, and yet, campus is most abuzz in the springtime. Perhaps it is the delivery from a long, harsh Minnesota winter. Perhaps students (Oles) are more clearly able to see the end of the semester. Or perhaps, love is in the air. Springtime is that last chance for seniors to meet the love of their life from a pool of hundreds of exceptional candidates.
The mentality described in the preceding narrative is often termed “Ring by Spring” by students and faculty at St. Olaf College. “Ring by Spring” encapsulates the idea that Oles get married to other Oles that they have met on campus. Our research was inspired by the fact that regardless of this myth, many students do not feel that they fit into the "Ring by Spring" stereotype. We intend to explore the degree to which Oles expect to meet their future spouse on campus and several variables that may influence such an expectation.
REVIEW OF LITERATURE
Marriage and the Transition to Adulthood
Over the last fifty years there have been numerous changes in the way young people understand adulthood and marriage. Historically, specific practices have marked the entrance to adulthood, such as finishing education, starting a job, getting married and becoming a parent (Arnett 1997; Carroll, Willoughby, Badger, Nelson, Barry, and Madsen 2007). By the late 1990s, Arnett (1997) found that neither college students nor people in their twenties placed much importance on marriage, parenting, finishing education, or purchasing a house (respectively 15%, 12%, 18%, and 14%) as markers of adulthood (Arnett 1997). Instead, young people today view adulthood more in terms of internal characteristics than the fulfillment of social norms and report that “accepting responsibility for the consequences of your actions” was the most important criterion for becoming an adult- 98% (Arnett 1997). All of these findings support the prevailing ideology of the United States, individualism.
Many scholars link these conceptual changes about adulthood to the rising average age at which people first marry. For the last half-century young people have continued to delay marriage. The U.S. Census has mapped this trend in both men and women. Between the years 1957 and 2005 the median age of first marriage for women has increased from 20.1 to 25.8 years and for men from 22.5 to 27.4 years (Carol et al. 2007). Despite the national tendency to delay marriage, the majority of young adults still place a high importance on getting married (Carol et al. 2007). Both men and women identified 25 years old as the ideal age of first marriage (Carol et al. 2007). This shows that perceived ideal age of first marriage is lower than the national average of actual first marriage.
Education and Career Goals
Education and career goals are linked as variables that influence marital timing. Economic factors appear to play a vital, although different, role in marital timing for men as compared to women. For women, increased education, growth of employment opportunity, and rise in labor-force participation have reduced their economic dependence on men, suggesting that women have increased agency to choose their marriage age (Sassler and Schoen 1999). Financially independent women often delay marriage until their late twenties and early thirties; however, there is no evidence to support a diminished desire to become married (Sassler and Schoen 1999).
In any case, financially independent and economically attractive men and women have a high probability of marriage. However, economically attractive men have a high probability of marriage between the ages of twenty and twenty-three, while economically attractive women have a high probability between the ages of twenty-eight and thirty-one (Sassler and Schoen 1999). More recent research shows that men increasingly prefer to marry women who have completed their education goals and are economically independent (Barnett et al. 2003). This research suggests that the increase in age at first marriage is due to the economic attitudes of both men and women.
The trend among both men and women to delay marriage until the mid to late twenties is also influenced by career and dual-career marriage concerns. The first ten years after college are both a time to launch careers and negotiate a marriage relationship. Barnett, Gareis, James, and Steele (2003) assert that the interplay between career and marital/family goals leads to concern about career-marital conflicts for college students. College seniors admit to having at least “a little” to “some” concern about these issues (Barnett et al. 2003). Barnett et al. (2003) suggest that young couples are choosing occupations with family-friendly hours or to postpone childbearing and have fewer children in order to combat family/career conflicts.
Gender Role Attitudes
Barber and Axim (1998) focus specifically on gender role attitudes and expectations about marital timing among young women. Although past studies have indicated that women with more traditional attitudes are more likely to marry at a young age, Barbers and Axim’s (1998) study indicates that early marriage rates are actually lower for women with traditional gender role attitudes if these women expect to attain high levels of education. Early marriage rates are higher for women who hold traditional gender role attitudes but do not expect to attain high levels of education (Barber and Axim 1998). Thus, the study found that women who experience high levels of dissonance between their educational aspirations and their marital plans have the lowest probability of early marriage (Barber and Axim 1998).
Similarly, Koball (2004) finds that the gender role attitudes of a young man can significantly influence his potential to marry and/or the age at which he marries. Koball (2004) found a strong association between level of income and rate of marriage for men who hold traditional gender role attitudes and cast themselves in the role of primary provider. Men who hold this attitude and have low levels of income are significantly less likely to marry than their higher earning counterparts. However, there is no correlation between income and rate of marriage for men who hold more egalitarian gender role attitudes (Koball 2004). This result suggests that timing of marriage may be influenced by the young man’s gender role attitudes and his perceived ability to fulfill said role.
Marriage Myths
Multiple socializing forces in our society, for example, the media, can create gaps in knowledge about marriage and produce marital myths. Studies by Larson (1988) and Ganong and Sharp (2000) found that there is dissonance between what college students think they know about marriage and what are actually just myths. Both studies delve deeper into the extent of students’ beliefs in marital myths and how these beliefs may affect their future marriage relationships.
Larson (1988) focused specifically on college students and their knowledge and/or perceptions about marriage. He saw that previous studies have identified an association between unsatisfied couples and couples who have unrealistic understandings of marriage. Thus, because inaccurate beliefs often affect how a person acts in a relationship, Larson (1988) believed it was important to gauge students’ beliefs in marital myths.
Both Larson (1988) and Ganong and Sharp (2000) evaluated participants’ romantic attitudes and found that students who believed fewer marital myths were females, students with less romantic attitudes and students who had completed a college level marriage and the family course. Our study draws upon these previous conceptualizations of romanticism to assess our participants’ romantic attitudes and investigate how they relate to marital timing.
After reviewing studies pertaining to marital timing, we found a gap in research concerning college students’ expectations towards marriage. Although several studies discussed college students briefly, we feel that this population warrants more direct attention. As undergraduate researchers at a college where marriage and the possibility of finding a spouse on campus is a common topic of discussion, we wondered to what degree undergraduate students actually expect to meet their spouse at college. At St. Olaf College there is a popular belief that many Oles in relationships with other Oles scramble to get engaged by the end of spring semester of senior year. Based on these interests, we decided to test the following hypotheses:
- Students who perceive that many Oles marry other Oles are more likely to expect to meet their spouse at St. Olaf.
- Students with more romantic attitudes toward marriage are more likely to expect to find their future spouse at St. Olaf
METHODS
Data for this study was collected using a survey questionnaire administered to 540 college students. The questions on the survey explored students’ attitudes towards marital timing and their expectations about meeting their spouse on their college campus. In order to investigate these issues we examined the relationship between two independent variables, romantic attitudes and student perceptions that Oles marry other Oles, and the dependent variable, students’ expectations to meet their spouse at St. Olaf.
Measurement
Romantic Attitudes. We measured romantic attitudes by using a Likert scale modeled on an index used in “The Marriage Quiz” (Larson 1988). In our study, participants were asked to indicate their level of agreement with several statements. We included three statements as “more romantic” (ex. “I believe in love at first sight”) and three statements as “less romantic” (ex. “Love is hard work”). Although we would have liked to expand this portion of our research, we used six statements because we were only allotted a small portion of a larger, joint survey. We constructed an index by assigning a number value to each response category (strongly agree, agree, disagree, and strongly disagree) and respondents were given a “romantic score”.
Perception that Oles marry other Oles. To measure this variable we asked students to indicate how common it is for St. Olaf students to marry other St. Olaf students. We asked, “In your opinion, how common is it for Oles to marry other Oles?” and provided the response categories of very common, somewhat common, somewhat uncommon, and very uncommon.
Student expectations to meet their spouse at St. Olaf. We measured the dependent variable by asking students to specify the degree to which they expect to meet their spouse at college. We asked, “Do you expect to meet your spouse at St. Olaf?” and provided the response categories of definitely, probably, probably not, and definitely not.
Validity and Reliability
In our study we achieved validity and reliability to ensure the credibility of our research.
Face Validity. Face validity is an agreement in the scientific community that the indicators measure what they propose to measure (Neuman 2007). In this study we achieved a degree of face validity by having our undergraduate peers and our advising professor review and critique our survey questions. However, we acknowledge that these questions have not been reviewed by the larger social-scientific community and eminent researchers in our field of study.
Content Validity. Content validity requires that the survey questions measure all aspects of the conceptual definition (Neuman 2007). We are confident that, after multiple revisions, our indicators measured the relationships that they were intended to measure. For example, we created a romantic index to measure students’ level of romanticism. We also asked respondents if they expected to meet their spouse at St. Olaf.
Criterion Validity. Criterion validity requires that the measures in our study be verified by an outside, independent party and have previously achieved success in other accepted studies (Neuman 2007). By modeling our questions after established and peer-reviewed studies we have achieved criterion validity. For our study we used a Likert scale to evaluate romantic attitudes. Countless studies have successfully employed the Likert scale to measure multiple indicators.
Reliability. Reliability means that the measure used will consistently measure the variable that it intends to. This is necessary to ascertain that data are legitimate (Neuman 2007). To increase the reliability of our study we took four steps suggested by Neuman: 1. We made sure that our objectives, hypotheses, and variables were clearly defined; 2. We created mutually exclusive and exhaustive response categories, which allow for precise measurement; 3. We utilized multiple indicators to ensure that we obtained a more holistic and accurate picture of the variables we set out to measure.; 4. We administered the survey questionnaire to our undergraduate peers to assess clarity, possible inconsistencies, or general flaws (Neuman 2007).
Sampling Procedures and Characteristics
Our target population was students at a small Midwestern college. Susan Canon, Director of Institutional Research, used a computer program to draw a simple random sample after excluding some categories of students per our request. We excluded students under the age of 18, students in our course, students who participated in our focus group, collaborating statistics fellows, students abroad, and part time students. These students were excluded due to previous exposure to research topics, absence from the college campus or vulnerability. Canon then gave us an email alias for the students in our sample so that we could reach them, without knowing their names. We asked 540 students to complete a survey and received a response rate of 58% for a total of 314 respondents. Our respondents were 25% first year students, 26% sophomore students, 23% juniors, 21% seniors, and 5% other; 62% were female and 38% were male. A large percentage of respondents (90%) listed their ethnicity as Non-Hispanic White.Out of 314 respondents, an overwhelming majority (304 students) reported that they were neither married nor engaged, while only 8 students were engaged and 2 students were married. All students were between age 18 and 24.
Ethics
We considered several ethical issues when administering the survey. First we addressed the issues of privacy. Because our survey was anonymous we did not have any identifying information to connect data to respondents, therefore their privacy could not be compromised. Secondly, respondents were given a cover letter that summarized the study, explained its voluntary nature and the potential risks involved, and provided information for respondents to give informed consent. And finally, before distributing the survey questionnaire, we submitted our proposal to the Institutional Review Board (IRB). The IRB issued its approval and confirmed that our study presents minimal risk to participants.
RESULTS
Hypothesis 1: Students who perceive that many Oles marry other Oles are more likely to expect to meet their spouse at St. Olaf.
Univariate frequency analyses revealed that over 60% of our respondents expect to definitely not or probably not find their future spouse at St. Olaf, while only 2% of students definitely expect to find their spouse at college (Figure 1). Despite the low percentage of students believing that they would not find a spouse at St. Olaf, the frequencies of responses for “Oles Marrying Oles” revealed that over 90% of respondents believed marriages between Oles are somewhat common or very common. Less than 7% of respondents believed it was somewhat uncommon or very uncommon for Oles to marry other Oles (Figure 2). Furthermore, respondents were asked to indicate what percentage of their classmates they believe get married to other Oles. The mean of the responses was 23%. However, the percentages given ranged between 0% and 80%. The standard deviation was 17, meaning that 68% of the percentages given by respondents were between 6% and 40% (Figures 3a and 3b).
Figure 1: Graph of Frequencies for Dependent Variable “Meet Spouse” at St. Olaf
Figure 2: Graph of Frequencies for Independent Variable “Oles Marrying Oles?”
Statistics |
||
Percentage Oles Marrying Oles |
||
N |
Valid |
297.00 |
Missing |
17.00 |
|
Mean |
23.18 |
|
Median |
20.00 |
|
Std. Deviation |
17.43 |
|
Figure 3a: Descriptive Statistics Table for “Percentage of Oles Marrying Oles”
Figure 3b: Graph for Frequencies for “Percentage of Oles Marrying Oles”
To determine the association between the perception of “Oles Marrying Oles” and the expectation to “Meet Spouse,” we ran a two tailed Spearman's rho correlation. The test found that there was no significant correlation between these variables (r (2) = 0.028, p > 0.05). The correlation coefficient equals 0.028 and has a significance level of 0.642, thus we cannot reject the null hypothesis (Table 1).
Table 1 Correlations: Meet Spouse at St. Olaf and Oles Marrying Oles |
||||
|
|
|
Meet Spouse at Olaf |
Oles Marrying Oles |
Spearman's rho |
Meet Spouse at Olaf |
Correlation Coefficient |
1.000 |
.028 |
Sig. (2-tailed) |
. |
.642 |
||
N |
278 |
275 |
||
Oles Marrying Oles |
Correlation Coefficient |
.028 |
1.000 |
|
Sig. (2-tailed) |
.642 |
. |
||
N |
275 |
310 |
||
In addition to our hypothesis, we explored the relationship between gender and the expectation to marry another St. Olaf student (Table 2). We are 95% confident that there is a gender difference in the expectation to marry another Ole, within a confidence interval of 0.024 to 0.356 (Table 3). The p-value is significant at 0.025 which would have allowed us to reject a null-hypothesis that gender and expectations to marry an Ole are not related.
Table 2 Group Statistics: Gender and Expect to Meet Spouse at St. Olaf |
|||||
|
Gender |
N |
Mean |
Std. Deviation |
Std. Error Mean |
Meet Spouse at Olaf |
male |
107 |
2.41 |
.713 |
.069 |
female |
167 |
2.22 |
.625 |
.048 |
|
Table 3 Independent Samples Test: Gender and Expect to Meet Spouse at St. Olaf |
||||||||||
|
|
Levene's Test for Equality of Variances |
t-test for Equality of Means |
|||||||
|
|
F |
Sig. |
t |
df |
Sig. (2-tailed) |
Mean Difference |
Std. Error Difference |
95% Confidence Interval of the Difference |
|
|
|
Lower |
Upper |
|||||||
Meet Spouse at Olaf |
Equal variances assumed |
6.859 |
.009 |
2.318 |
272 |
.021 |
.190 |
.082 |
.029 |
.351 |
Equal variances not assumed |
|
|
2.252 |
204.345 |
.025 |
.190 |
.084 |
.024 |
.356 |
|
Hypothesis 2: Students with more romantic attitudes toward marriage are more likely to expect to find their future spouse at St. Olaf.
Univariate data analysis found a normal distribution for the romantic indexes (M =13, SD = 3, Range = 17 – 21) (Figure 4). A two-tailed Spearman's rho test found a weak, positive correlation between romantic attitudes and expectation to meet spouse at St. Olaf, with a correlation coefficient of 0.171 and a p-value of 0.005. The p-value is statistically significant so we can reject the null-hypothesis that there is no correlation between romantic attitudes and expectation to meet spouse at St. Olaf (Table 4).
Figure 4: Graph of Romantic Index Scores
Table 4 Correlations: Romantic Index and Meeting Spouse at St. Olaf |
||||
|
|
|
RomanticIndex |
BinSpouseSTO |
Spearman's rho |
RomanticIndex |
Correlation Coefficient |
1.000 |
.171 ** |
Sig. (2-tailed) |
. |
.005 |
||
N |
303 |
269 |
||
BinSpouseSTO |
Correlation Coefficient |
.171 ** |
1.000 |
|
Sig. (2-tailed) |
.005 |
. |
||
N |
269 |
278 |
||
**. Correlation is significant at the 0.01 level (2-tailed). |
|
|
||
An independent samples t-test comparing the means of the romantic index by gender demonstrated that men and women have similar romantic attitudes (male M = 13.09, female M = 13.06) (Tables 5a and 5b). Men expect to marry at a slightly younger age, 26.3, than the national average, 27.4. Women expect to marry at approximately the same age, 26.1, as the national average, 25.8 (Carroll et. al. 2007) (Tables 6a and 6b).
Table 5a Group Statistics: Romantic Index and Gender |
|||||
|
Gender |
N |
Mean |
Std. Deviation |
Std. Error Mean |
RomanticIndex |
male |
117 |
13.09 |
2.720 |
.251 |
female |
185 |
13.06 |
2.507 |
.184 |
|
Table 5b Independent Samples Test: Romantic Index and Gender |
||||||||||
|
|
Levene's Test for Equality of Variances |
t-test for Equality of Means |
|||||||
|
|
F |
Sig. |
t |
df |
Sig. (2-tailed) |
Mean Difference |
Std. Error Difference |
95% Confidence Interval of the Difference |
|
|
|
Lower |
Upper |
|||||||
RomanticIndex |
Equal variances assumed |
.855 |
.356 |
.113 |
300 |
.910 |
.035 |
.306 |
-.568 |
.637 |
Equal variances not assumed |
|
|
.111 |
231.942 |
.912 |
.035 |
.312 |
-.580 |
.649 |
|
Table 6a Group Statistics: Gender and Ideal Age of Marriage |
|||||
|
Gender |
N |
Mean |
Std. Deviation |
Std. Error Mean |
Ideal Marriage Age |
male |
94 |
26.27 |
2.156 |
.222 |
female |
157 |
26.11 |
2.262 |
.180 |
|
Table 6b Independent Samples Test: Gender and Ideal Marriage Age |
||||||||||
|
|
Levene's Test for Equality of Variances |
t-test for Equality of Means |
|||||||
|
|
F |
Sig. |
t |
df |
Sig. (2-tailed) |
Mean Difference |
Std. Error Difference |
95% Confidence Interval of the Difference |
|
|
|
Lower |
Upper |
|||||||
Ideal Marriage Age |
Equal variances assumed |
.030 |
.863 |
.522 |
249 |
.602 |
.151 |
.290 |
-.420 |
.722 |
Equal variances not assumed |
|
|
.528 |
203.292 |
.598 |
.151 |
.286 |
-.413 |
.716 |
|
In conclusion, bivariate statistical analysis of our first hypothesis showed no significant relationship between students who perceive that Oles marry other Oles and students who expect to m eet their spouse at St. Olaf. In our second hypothesis, bivariate statistical analysis confirmed that there is a weak, positive and statistically significant association between students’ romantic attitudes and their expectations to meet their spouse at St. Olaf.
DISCUSSION
Our first hypothesis was not supported by the results of the correlation test and, therefore, we cannot conclude that there is an association between a student’s perception about Oles marrying other Oles and their own expectation to meet their spouse in college. One reason that may explain the lack of association between these two variables is the fact that students might underreport their expectations to marry another St. Olaf student. Even with an anonymous survey, students might not want to admit, even to themselves, that they fit the "Ring by Spring" stereotype. Since research shows that the national age of first marriage is increasing, it may be less socially acceptable to expect to marry at a younger age (Caroll et al. 2007). Therefore, the “Ring by Spring” phenomenon may viewed negatively by the student body, and thus, students might feel that there is a stigma associated with expecting to meet their spouse at St. Olaf. Another reason for this insignificant association could be that students at St. Olaf are more focused on their future career and education goals than on meeting and marrying a college sweetheart. This speculation is supported by Caroll et al.’s (2007) finding that students who placed higher priority on career and education development tended to postpone marriage.
The second hypothesis showed a weak, positive correlation between students’ romantic attitude scores and their expectation to meet their spouse at St. Olaf. Thus, students with higher romantic scores were more likely to expect to meet their spouse in college than were their less romantic counterparts. A student with a more romantic attitude might be more likely to idealize marriage and thus be enchanted by the idea of “Ring by Spring.” This could contribute to their expectation to be “swept off their feet” by a college romance. The correlation might have been stronger if we had been able to add more statements to the romantic index because a score based on more aspects of romanticism would be a more accurate measure.
While analyzing our romantic attitudes index, we observed several additional findings regarding gender. First, although social stereotypes might cause a person to expect women to have more romantic attitudes than men, we found that men and women at St. Olaf have approximately the same average score on the romantic index. Secondly, the data showed that, on average, men and women from St. Olaf have nearly the same expected ideal age of first marriage. One reason for these findings could be that in recent years, society has begun to place less emphasis on dichotomized gender roles. Larson (1988) points out that, in the past, women were socialized to think about marriage and their future spousal role far more often than were men. Our research could suggest that today male and female college students are receiving more equal socialization regarding marital expectations. Also, many recent studies show that women are beginning to place the same emphasis as men on future career and educational goals, so it is not surprising that both genders would also have similar ideal marriage ages (Sassler and Schoen 1998).
CONCLUSION
In this study we endeavored to discover how romantic attitudes and the perception that Oles marry other Oles may influence the degree to which college students expect to find their spouse at college. We found that students who had higher romantic attitude scores were more likely to expect to meet their spouse at St. Olaf than those with lower romantic scores. However, we discovered that students who perceive that many Oles marry other Oles do not necessarily have a high expectation to meet their spouse at St. Olaf.
The social science community has conducted little research regarding college students’ marital expectations. Several studies have investigated the influence of gender role attitudes and career and education goals on expected age of first marriage; however, more research needs to explore the topic of college students and the expectation to find their spouse while in college. Although college may seem to be an ideal opportunity to meet a spouse due the close proximity of thousands of young adults, the results of our study imply otherwise. Despite widespread beliefs that St. Olaf students have a propensity to marry other St. Olaf students, few students actually believe that they, themselves, will meet their future spouse on campus.
One weakness of our study was the limited survey space. For example, we had hoped to include more statements on our Likert scale in order to produce a more accurate romantic index score. Another limitation of our study was that most of our questions asked about future expectations and it can be difficult to predict one’s hypothetical actions. However, we could not avoid asking about future expectations due to the fact that an overwhelming majority of our population is currently unmarried. One strength of our electronic study survey design was that it was completely anonymous and it generated a high response rate. Our research also took an important step toward understanding how college students view marriage. Since the research on college students’ expectations about marriage is sparse, our study might be used as background data for future research in this area. By avoiding our weakness and building on our strengths, future studies could produce more pertinent data on this topic.
Appendix: Additional Analysis for Ring by Spring?: Attitudes Toward Marriage Among College Students
Kate Virkler, St. Olaf College Center for Interdisciplinary Research
Further analysis was conducted on the data after the final paper on “Ring by Spring?: Attitudes Toward Marriage Among College Students” had been completed.
In the paper, the second hypothesis states that people with more romantic attitudes will be more likely to expect to meet their future spouse at St. Olaf. The nonparametric Spearman’s correlation gave statistically significant results. However, since the romantic attitude index is approximately normally distributed, we can use more specific parametric statistical tests to further support this hypothesis. Using a logistic regression with the expectation to meet one’s spouse at St. Olaf as the binary response and the romantic index as the predictor, we can see that every point increase on the romantic index results in a 17.2% increase (95% CI [5.9%,29.7%]) in the odds of expecting to meet one’s spouse at St. Olaf. This relationship is highly significant (z-statistic of 9.399 and p-value of .002), and the hypothesis that people with more romantic attitudes will be more likely to expect to meet their spouse at St. Olaf is further supported.
Variables |
β |
S.E |
z |
d.f. |
Sig(p) |
Exp(β) |
95%C.I. |
Exp(β) |
Constant |
-2.669 |
.702 |
14.447 |
1 |
.000 |
.069 |
|
|
Romantic Index |
.159 |
.052 |
9.399 |
1 |
.002 |
1.172 |
1.059 |
1.297 |
We were also interested in investigating whether or not class year had an effect on one’s expectations to marry someone from St. Olaf. We decided to add the variable of class year into the logistic regression model above to test this. The class years used were first years, sophomores, juniors and seniors. Those that listed “other” as their class year were excluded. Holding class year constant, we can see that the romantic index variable is still significant (p=.005) and a one unit increase in the romantic index results in a 16.8% increase (95% CI [4.8,30.2]) in the odds of expecting to meet one’s spouse at St. Olaf. The baseline for our class year variable is seniors. The only class year that significantly differed from seniors were first years (p=.034) who had a 137% increase in the odds of expecting to meet his or her spouse at St. Olaf (95% CI [6.9,426.4]) after adjusting for differences in romantic index. Juniors also seemed to have slightly higher expectations with an estimated increase of 99.5% in the odds of expecting to meet one’s spouse, however the p-value=.081, therefore the results are not significant.
Variables |
β |
S.E |
z |
d.f. |
Sig(p) |
Exp(β) |
95%C.I. |
Exp(β) |
Constant |
-3.065 |
.800 |
14.691 |
1 |
.000 |
.047 |
|
|
Romantic Index |
.156 |
.055 |
7.875 |
1 |
.005 |
1.168 |
1.048 |
1.302 |
Class Year |
|
|
6.194 |
3 |
.103 |
|
|
|
Junior |
.691 |
.396 |
3.042 |
1 |
.081 |
1.995 |
.918 |
4.337 |
Sophomore |
.232 |
.400 |
.337 |
1 |
.561 |
1.261 |
.576 |
2.762 |
First Year |
.864 |
.407 |
4.508 |
1 |
.034 |
2.372 |
1.069 |
5.264 |
We also tested the relationship between class year and expectations to meet a spouse at St. Olaf without adjustment for romantic index differences by a Chi-square test. The Pearson Chi-square statistic resulted in a value of 7.3 with a p-value of .063. While this is not a statistically significant finding at the .05 level, it is still interesting to note that it’s plausible that with more data there may be an association between class year and expectations to meet your future spouse.
Class Year |
Will not meet spouse at St. Olaf |
Will meet spouse at St. Olaf |
Total |
First Year |
37 (54%) |
32 (46%) |
69 |
Sophomore |
51 (68%) |
24 (32%) |
75 |
Junior |
35 (56%) |
27 (44%) |
62 |
Senior |
42 (74%) |
15 (26%) |
57 |
Total |
165 (63%) |
98 (37%) |
263 |
Chi-Square Test |
Value |
d.f. |
Asymp. Sig (2-sided) |
Pearson Chi-Square |
7.310 |
3 |
.063 |
We were also interested in investigating whether or not the number of hookups someone has is related to their ideal age of marriage. We performed a linear regression analysis to test this hypothesis with the ideal age of marriage as the response variable and the number of hookups as the explanatory variable. First, the regression was done excluding anyone who had no history of hookups at St. Olaf. There were no significant results (t=-0.27, p=.785).
Variables |
β |
S.E |
t |
Sig(p) |
95% |
C.I. |
Constant |
26.421 |
.317 |
83.400 |
.000 |
25.793 |
27.050 |
Total No. of Hookups |
-.009 |
.034 |
-.273 |
.785 |
-.076 |
.057 |
The analysis above was repeated with all respondents; again, there was no significant relationship between number of hookups and ideal age of marriage (t=0.64, p=.520).
Variables |
β |
S.E |
t |
Sig(p) |
95% |
C.I. |
Constant |
26.060 |
.152 |
171.324 |
.000 |
25.760 |
26.359 |
Total No. of Hookups (all) |
.016 |
.025 |
.644 |
.520 |
-.034 |
.067 |
In conclusion, the additional analysis has further supported the hypothesis in the paper that people with more romantic attitudes will be more likely to meet their future spouse at St. Olaf. Additionally, class year could be a factor in someone’s expectations to meet their future spouse at St. Olaf.
SOURCES CITED
Arnett, Jeffrey J. 1997. “Young People’s Conception of the Transition to Adulthood.” Youth & Society 29: 3-23.
Barber, Jennifer S. and William G. Axim. 1998. “Gender Roles Attitudes and Marriage Among Young Women.” The Sociological Quarterly 39: 11-29.
Barich, Rachel Rosemand; and Denise D. Bielby. 1996. “Rethinking Marriage: Change and Stability in Expectations, 1967-1994.” Journal of Family Issues 17: 139-169.
Barnett, Rosalind Chait, Karen C. Gareis, Jacquelyn Boone James, and Jennifer Steele. 2003. “Planning Ahead: College Seniors’ Concerns About Career–Marriage Conflict.” Journal of Vocational Behavior 62: 305-319.
Carroll, Jason S, Brian Willoughby, Sarah Badger, Larry J. Nelson, Carolyn McNamara Barry, and Stephanie D. Madsen. 2007. “So Close, Yet So Far Away: The Impact of Varying Marital Horizons on Emerging Adulthood.” Journal of Adolescent Research 22: 219-47.
Ganong, Lawrence and Elizabeth Sharp. 2000. "Raising Awareness about Marital Expectations: Are Unrealistic Beliefs Changed by Integrative Teaching?" National Council on Family Relations 49: 71-76.
Larson, Jeffery H. 1988. "The Marriage Quiz: Students' Beliefs in Selected Myths about Marriage." Family Relations 37: 3-11.
Koball, Heather L. 2004. “Crossing the Threshold: Men’s Incomes, Attitudes Toward the Provider Role, and Marital Timing.” Sex Roles 51: 387-395.
Sassler, Sharon and Robert Schoen. 1999. “The Effect of Attitudes and Economic Activity on Marriage.” Journal of Marriage and the Family 61: 147-159.
Acknowledgements: Kate Virkler, Consulting Statistics Fellow, Center for Interdisciplinary Research, St. Olaf College; Ryan Sheppard, Supervising Professor
