PBA Home > Institutional Research & Analysis > Surveys > Senior Survey 2008 > Satisfaction Correlates
Senior Survey, Spring 2008
Correlates of Seniors' Satisfaction with the Undergraduate Experience
Summary of Results. (In addition to the summary, you may wish to view the full report.)
In the Senior Survey, students were asked about their satisfaction pertaining to three facets of their undergraduate experience:
- Overall experience
- Academic experience
- Academic experience in their respective majors
2,695 seniors responded, with a response rate of 40%. The three satisfaction measures are moderately related to one another, with correlations ranging from .51 to .64.
We investigated groups of explanatory variables, shown in the table below, to determine how they relate to the satisfaction measures. These include characteristics of students at entry to CU-Boulder (e.g., gender, race/ethnicity), characteristics in the CU-Boulder entry year (e.g., in honors courses, in the Presidential Leadership Class), characteristics in the term surveyed (e.g., Colorado resident versus out-of-state student, cumulative GPA at the end of the prior term), and perceptions of their respective majors as reported on the Senior Survey (e.g., amount of structure in the major, difficulty of courses in the major).
Results: The available variables provide only a very limited explanation of variation in students' self-reported levels of satisfaction. The percentage of explained variance in satisfaction ranged from 3% for overall satisfaction to 8% for both satisfaction with academic experience and satisfaction with academic experience in a student's major.
Higher cumulative GPAs are related to higher levels of student satisfaction. This result does not, of course, indicate that high GPAs cause high satisfaction or that low GPAs cause dissatisfaction. Rather, they indicate a positive association between GPA and satisfaction.
Range of satisfaction: Over half of seniors selected the "satisfied" option on the "overall experience" item; 81% reported being either satisfied or very satisfied, with the remainder reporting that they were neutral concerning their experience (13%), dissatisfied (5%), or very dissatisfied (1%). Distributions on the other satisfaction items are similar.
The low proportions selecting the "dissatisfied" options probably reflect
- departure from CU-Boulder of seriously dissatisfied students before attaining senior status
- lower chances of responding to the survey among those dissatisfied
- true high satisfaction
- a tendency among respondents to any survey to cluster on the "satisfied" end of any scale.
 |
The relationship of satisfaction to other information about the student
| Explanatory Variables Examined, with Results |
| At entry to CU-Boulder |
| Gender |
| Race/ethnicity |
| First generation college student |
| SAT math score |
| SAT verbal score |
| ACT composite score |
| Predicted CU GPA (from HS GPA and test scores) (PGPA) |
| Cumulative transfer hours counted toward degree |
| Results: The variables in this group, alone and in combination, account for miniscule portions of |
| variance in seniors' satisfaction. There is a statistically reliable but small positive relationship |
| between satisfaction with the academic experience (not the overall experience) and |
| PGPA, which was investigated as a control variable in some later steps. |
| In the CU-Boulder entry year |
| In a Residential Academic Program |
| In the Presidential Leadership Class |
| In honors courses |
| Norlin Scholar and/or recipient of the Boettcher scholarship |
| Intercollegiate athlete |
| Results: None of these entry year experiences, alone or in combination, are notably related |
| to senior satisfaction, measured with PGPA controlled or not. |
| Responses on prior CU-Boulder surveys |
| Fewer than 100 senior survey respondents had responded to either the 2006 climate survey or |
| the 2006 NSSE survey. Therefore we did not attempt checks on relationships of the earlier |
| responses to senior satisfaction. |
| In the term surveyed |
| Cumulative GPA at end of the prior term |
| Full- vs. part-time CU-Boulder student |
| Colorado resident vs. out of state student |
| College |
| Results: Higher cumulative GPAs are associated with higher levels of self-reported satisfaction. |
| Cumulative GPA is the only variable related to senior satisfaction consistently and with |
| high statistical significance. It relates to satisfaction even after controlling for PGPA. |
| The plots below illustrate the relationship for satisfaction with the overall experience at |
| CU-Boulder. These results do not, of course, indicate that high GPAs cause high satisfaction |
| or that low GPAs cause dissatisfaction. Rather, they indicate a positive association between |
| GPA and self-reported satisfaction. |
| Perceptions of the major on the senior survey |
| Amount of structure (flexible to rigid) |
| Difficulty of courses (easy to difficult) |
| Degree of emphasis or focus (broad to specialized) |
| Results: Perceptions of more specialization and more structure in the major are related |
| to satisfaction, especially with the academic experience in the major -- but the |
| relationships are weak, with or without controlling for cumulative GPA. In addition, |
| students responding "about right" -- the midpoint of all three scales -- were somewhat |
| more satisfied than those answering on either end of the scales. |
| Demographics controlling for cumulative GPA, and interactions |
| Gender, race/ethnicity, residency, and first generation student |
| Results: Females and non-residents report slightly higher satisfaction (0.1 - 0.2 on a |
| 5-point scale), on average, after controlling for cumulative GPA; race/ethnicity and |
| first generation status are unrelated to satisfaction. No interactions among the |
| demographic variables, or with cumulative GPA, added to the explanations. |
| All explanatory variables combined |
| Results: The available variables provide only a very limited explanation of variation in |
| students' self-reported levels of satisfaction. The percentage of explained variance in |
| satisfaction ranged from 3% for overall satisfaction to 8% for both satisfaction with academic |
| experience and satisfaction with academic experience in a student's major. |
| The inability of the available variables to explain sufficiently students' self-reported |
| satisfaction suggests that variation in satisfaction might be better explained by variables |
| about which we have little or no information. Examples of such variables are social |
| relationships with peers, social activities, and interaction with faculty members. |
Visualizing the relationship of satisfaction with cumulative GPA: As noted above, students with higher cumulative GPAs are somewhat more satisfied on all dimensions; the relationship is statistically reliable but weak. No other variables examined show stronger relationships.
The plots below illustrate the relationship between GPA and satisfaction with the overall experience at CU-Boulder. The bar near the center of each box denotes the median GPA for a particular level of overall satisfaction. The 1% of students who reported being very dissatisfied with their overall experience have a lower median GPA than do the 25% who report being very satisfied (2.9 vs. 3.3, respectively). There are only very small differences (about 0.1) in median GPA among students who report being dissatisfied, neutral, satisfied, or very satisfied. These results do not, of course, indicate that high GPAs cause high satisfaction or that low GPAs cause dissatisfaction. Rather, they indicate a positive association between GPA and self-reported satisfaction. There are almost certainly elements of three sorts of causality operating:
- academic success and a higher cumulative GPA leads to satisfaction
- satisfaction with CU-Boulder leads to academic success and a higher cumulative GPA
- one or more external factors leads to both satisfaction and academic success. These might include student motivation and interactions with peers and faculty.
It is worth noting that senior survey non-respondents have slightly lower GPAs than do respondents (averages of 3.0 and 3.2, respectively). The relationship between GPA and satisfaction might differ for non-respondents, and non-respondents might be less satisfied.
 |
Last revision 02/25/09
PBA Home | Strategic Planning | Institutional Research & Analysis |
Budget & Finances | Questions? Comments?
15 UCB, University of Colorado Boulder, Boulder, CO 80309-0015, (303)492-8631
© 2001, The Regents of the University of Colorado