Julian J. Meimban III
Edwin A. Benigno
Luzale D.Henson
Audrey Lyle dM. Diego
New Era University
How to Cite:
Meimban III, J. J., Benigno, E. A., Henson, L. D., & Diego, A. L. dM. (2024). Satisfaction survey of NEU’s science fair. NEU Knowledge Journal: A Compilation of Researches of New Era University Faculty, Staff, Students, and Administrators, 1(1), 27-36. https://doi.org/10.64303/kj-urc-neu-SfUeNss
ABSTRACT
Satisfaction survey of a Science Fair was examined using Two-Way ANOVA. The dependent variable (AVG) indicated a respondent’s mean score on a 7-item, 5-point Likert scale (1-Very Poor, 2-Poor, 3-Satisfactory, 4-Good, 5-Excellent). The seven items rated were Venue/place (VENUE), Inventiveness (INVENT), Commercial Viability (IMPACT), Educational Value (EDUC), Audio-Visual Presentation (PRESENT), Organization (ORGANIZE), and Overall Experience/Satisfaction (OVERALL). The independent variables were GENDER (MALE, FEMALE) and GROUP (STUDENT – student, EXHIB – exhibitor, FAC-faculty). The survey instrument was distributed to 159 randomly selected respondents, based on power analysis as recommended by Cohen (1988). The survey yielded 95% response rate and responses had an estimated internal consistency reliability of .92 (Cronbach’s Alpha). On the average, Science Fair attendees expressed satisfaction beyond satisfactory level (M = 4.16), p < .001, effect size (d) = 1.53 (> large according to Cohen, 1988). Profile plots of marginal means revealed interaction between GROUP and GENDER. Post hoc test using Games-Howell showed that female faculty were significantly less satisfied (M = 3.74, SD = .88) than female student (M = 4.47, SD = .40), p = .012, 95% CI (.12, 1.34) and also than female exhibitor (M = 4.46, SD = .47), p = .017, 95% CI (.09, 1.35). Other GROUP-GENDER pairwise comparisons were found not statistically significantly different (p > .05) in satisfaction level (beyond Satisfactory).
Keywords: Satisfaction rating, Factorial ANOVA, Games-Howell, profile plot, post hoc
INTRODUCTION
This study was conducted to assess the satisfaction ratings of attendees in the NEU’s 2017 Science Fair held on September 7 to 9, 2017 at the Professional Schools Building, New Era University, No. 9 Central Avenue, New Era, Quezon City. Researchers were interested to know the satisfaction rating of attendees on some facets of the Science Fair such as venue, presentation, and educational impact. Assessment procedure followed in this study involved more rigorous consideration on experimental design, hypothesis testing, and statistical test as compared to previous evaluation studies on the Science Fair. Findings of this research will provide organizers of the next Science Fair scientifically grounded insights on how to make the next annual event even more educationally captivating and intellectually rewarding.
Research Questions
Research Question 1: What is the average satisfaction rating of attendees?
Ho: Average satisfaction rating is 3 or Satisfactory on the scale (μ=3).
Ha: μ is not equal to 3.
Research Question 2: Are the satisfaction ratings among groups (students, exhibitors, and faculty) significantly different?
Ho: Satisfaction ratings of groups are the same (μ1 = μ2 = μ3).
Ha: Not all of the μs are equal.
Research Question 3: Are the satisfaction ratings between male and female significantly different?
Ho: Satisfaction ratings of male and female are the same (μ1 = μ2).
Ha: μ1 is not equal to μ2.
Research Question 4: Do the differences in the satisfaction ratings among groups vary depending on gender?
Ho: The differences in the satisfaction ratings among groups remain constant for male and female.
Ha: The differences in the satisfaction ratings among groups vary for male and female.
METHODOLOGY
Attendees were classified into groups (students, exhibitors, and faculty) and gender (male and female). A 7-item, 5-point Likert scale was developed and used as instrument to measure a respondent’s satisfaction level. The study was analyzed using two-way ANOVA.
Sample Size
Sample size was calculated based on 2×3 ANOVA design involving three groups, two gender, .80 power, .25 effect size (medium) for the main effects and interaction effect, and .05 significance level. For this specification, total sample size of 159 (53 cases per group) was required as recommended by Cohen (1988). And for each group, 26 were male and 27 female.
Survey Questionnaire and Distribution
The 7-item questionnaire developed concerned about (1) venue/place (VENUE), (2) inventiveness (INVENT), (3) commercial viability (IMPACT), 4) educational value (EDUC), (5) audio-visual presentation (PRESENT), (6) organization (ORG), and 7) overall experience/satisfaction (OVERALL). Respondents were asked to rate each item using a 5-point Likert scale, where: 5 – Excellent, 4 – Good, 3 – Satisfactory, 2 – Poor, and 1 – Very Poor. Six student-volunteers were trained, instructed, and tasked to administer the distribution of questionnaire. Two student-volunteers were assigned to survey each group. One student was tasked to survey male respondents; the other student, female respondents. Surveyors were instructed to select respondents randomly from among those who had gone to the Science Fair.
They were further instructed not to choose friends or relatives and make sure that respondents filled out the questionnaire independently and truthfully.
Validity and Reliability
Before it was used, the questionnaire underwent validity and reliability testing. For content validity, the questionnaire was reviewed and rectified by three test construction professionals. For reliability test, the estimated internal consistency reliability (Cronbach’s Alpha) was .92, greater than the .70 recommended minimum acceptable value for responses to be considered
reliable or consistent (Bohrnstedt and Knoke, 1988).
Inter-correlation of items ranged from .48 to .78, adjudged moderate to strong relationship (Salkind, 2000). With high estimate of internal consistency reliability and inter-correlation of items, a summated or composite score was computed and used as an index of an overall rating. A summated variable (SCORE) was computed as follows: SCORE = VENUE + INVENT + IMPACT + EDUC + PRESENT + ORG + OVERALL. Variable AVG was designated and computed as the mean of SCORE.
RESULTS AND DISCUSSION
Respondents
Out of the 159 survey questionnaires distributed, 151 were returned (95.0% response rate). But out of 151, only 143 were found usable (94.7%) because some forms had missing entries. By group, faculty (FAC) had the least usable number of cases (26.6%). By gender, female (52.4%) and male (47.6%) had comparable number of usable cases (Table 1).
Moreover, AVG for various combinations of group and gender appears greater for STUDENT and for EXHIB than that for FAC (Table 2). However, whether or not this apparent difference was significant or not was examined in the hypothesis testing.
Research Question 1
One-Sample t-test (α = .05) for AVG was statistically significant, t(145) = 18.46, p < .001), effect size (d) = 1.53 (> large according to Cohen, 1988). In other words, the sample mean (4.16) and the hypothesized mean (3, Satisfactory on the scale) were statistically significantly different. This result suggests that, on the average, Science Fair attendees expressed satisfaction beyond Satisfactory level (Table 2).
Research Questions 2, 3, and 4 will be addressed together as they are conceptually and statistically interrelated.
ANOVA of interaction effect showed that GROUP x GENDER was not statistically significant, F(2,132) = 1.091, p = .339, observed power = .24 (Table 3). Test for main effects showed that GENDER was not statistically significant, F(1,132) = 1.235, p = .268, observed power = .20. On the other hand, GROUP was found statistically significant, F(2,132) = 6.690, p = .002, effect size (partial η) = .30 (medium), observed power = .91. These findings, however, should be interpreted with caution because the equal variance assumption for ANOVA test was violated as shown by significant Levene’s test, F(5,132) = 3.101, p = .011. In fact, further investigation disclosed that the plot of marginal means for FEMALE intersects with that for MALE at the FAC group (see Exhibit 1). This finding, which contradicts the result of statistical test for interaction effect, suggests that there could be a significant GROUP and GENDER interaction which apparently was not detected because of insufficient observed power for the test of interaction effect (.24). Indeed, further statistical test revealed significant interaction between GROUP and GENDER as indicated by the variable cellcode (Table 4).
Omnibus ANOVA for the variable cellcode, which specifies various combination of the levels of GROUP and GENDER, was statistically significant F(5,132) = 3.730, p = .003, effect size (partial η) = .35 (large), observed power = .93 (Table 4). Multiple comparisons, using Games- Howell following significant Levene’s test, F(5,132) = 3.101, p = .011, revealed statistically significant differences between cellcodes. In particular, cellcode 6 (FAC,FEMALE) (M = 3.74, SD = .88) expressed a degree of satisfaction significantly lower than that given by cellcode 2 (STUD,FEMALE) (M = 4.47, SD = .40), p = .012, 95% CI (.12, 1.34) and also by cellcode 4 (EXHIB,FEMALE) (M = 4.46, SD = .471), p = .017, 95% CI (.09, 1.35) (Table 5 and Table 6).
Other cellcode pairwise comparisons were found not statistically significantly different, meaning, attendees described by the cellcodes being compared expressed similar degree of satisfaction (> 3 or beyond Satisfactory) (Table 5).
BIBLIOGRAPHY
Bohrnstedt, George W. and Knoke, David. 1988. Statistics for Social Data Analysis. 2nd Edition. USA: F. E. Peacock Publishers, Inc.
Cohen, Jacob. 1988. Statistical Power Analysis for the Behavioral Sciences. 2nd Edition. USA: Lawrence Erlbaum Associates, Inc.
Salkind, Neil J. 2000. Exploring Research. 4th Edition. USA:Prentice-Hall, Inc.
SPSS v23.
