| ANOVA | |||||
| R’s socioeconomic index (2010) | |||||
| Sum of Squares | df | Mean Square | F | Sig. | |
| Between Groups | 474058.682 | 4 | 118514.671 | 385.981 | .000 |
| Within Groups | 743670.506 | 2422 | 307.048 | ||
| Total | 1217729.188 | 2426 |
| Test of Homogeneity of Variances | |||
| R’s socioeconomic index (2010) | |||
| Levene Statistic | df1 | df2 | Sig. |
| 23.938 | 4 | 2422 | .000 |
| Multiple Comparisons | |||||||
| Dependent Variable: R’s socioeconomic index (2010) | |||||||
| (I) RS HIGHEST DEGREE | (J) RS HIGHEST DEGREE | Mean Difference (I-J) | Std. Error | Sig. | 95% Confidence Interval | ||
| Lower Bound | Upper Bound | ||||||
| Bonferroni | LT HIGH SCHOOL | HIGH SCHOOL | -10.6923* | 1.1432 | .000 | -13.904 | -7.480 |
| JUNIOR COLLEGE | -19.9879* | 1.6560 | .000 | -24.640 | -15.335 | ||
| BACHELOR | -32.4796* | 1.3119 | .000 | -36.166 | -28.794 | ||
| GRADUATE | -46.2190* | 1.4764 | .000 | -50.367 | -42.071 | ||
| HIGH SCHOOL | LT HIGH SCHOOL | 10.6923* | 1.1432 | .000 | 7.480 | 13.904 | |
| JUNIOR COLLEGE | -9.2956* | 1.3924 | .000 | -13.208 | -5.383 | ||
| BACHELOR | -21.7874* | .9580 | .000 | -24.479 | -19.096 | ||
| GRADUATE | -35.5267* | 1.1732 | .000 | -38.823 | -32.230 | ||
| JUNIOR COLLEGE | LT HIGH SCHOOL | 19.9879* | 1.6560 | .000 | 15.335 | 24.640 | |
| HIGH SCHOOL | 9.2956* | 1.3924 | .000 | 5.383 | 13.208 | ||
| BACHELOR | -12.4918* | 1.5340 | .000 | -16.802 | -8.182 | ||
| GRADUATE | -26.2311* | 1.6768 | .000 | -30.942 | -21.520 | ||
| BACHELOR | LT HIGH SCHOOL | 32.4796* | 1.3119 | .000 | 28.794 | 36.166 | |
| HIGH SCHOOL | 21.7874* | .9580 | .000 | 19.096 | 24.479 | ||
| JUNIOR COLLEGE | 12.4918* | 1.5340 | .000 | 8.182 | 16.802 | ||
| GRADUATE | -13.7394* | 1.3382 | .000 | -17.499 | -9.980 | ||
| GRADUATE | LT HIGH SCHOOL | 46.2190* | 1.4764 | .000 | 42.071 | 50.367 | |
| HIGH SCHOOL | 35.5267* | 1.1732 | .000 | 32.230 | 38.823 | ||
| JUNIOR COLLEGE | 26.2311* | 1.6768 | .000 | 21.520 | 30.942 | ||
| BACHELOR | 13.7394* | 1.3382 | .000 | 9.980 | 17.499 | ||
| Games-Howell | LT HIGH SCHOOL | HIGH SCHOOL | -10.6923* | .9292 | .000 | -13.235 | -8.149 |
| JUNIOR COLLEGE | -19.9879* | 1.6288 | .000 | -24.459 | -15.517 | ||
| BACHELOR | -32.4796* | 1.2173 | .000 | -35.808 | -29.151 | ||
| GRADUATE | -46.2190* | 1.2495 | .000 | -49.639 | -42.799 | ||
| HIGH SCHOOL | LT HIGH SCHOOL | 10.6923* | .9292 | .000 | 8.149 | 13.235 | |
| JUNIOR COLLEGE | -9.2956* | 1.5125 | .000 | -13.455 | -5.137 | ||
| BACHELOR | -21.7874* | 1.0566 | .000 | -24.677 | -18.898 | ||
| GRADUATE | -35.5267* | 1.0936 | .000 | -38.523 | -32.531 | ||
| JUNIOR COLLEGE | LT HIGH SCHOOL | 19.9879* | 1.6288 | .000 | 15.517 | 24.459 | |
| HIGH SCHOOL | 9.2956* | 1.5125 | .000 | 5.137 | 13.455 | ||
| BACHELOR | -12.4918* | 1.7047 | .000 | -17.167 | -7.817 | ||
| GRADUATE | -26.2311* | 1.7279 | .000 | -30.970 | -21.492 | ||
| BACHELOR | LT HIGH SCHOOL | 32.4796* | 1.2173 | .000 | 29.151 | 35.808 | |
| HIGH SCHOOL | 21.7874* | 1.0566 | .000 | 18.898 | 24.677 | ||
| JUNIOR COLLEGE | 12.4918* | 1.7047 | .000 | 7.817 | 17.167 | ||
| GRADUATE | -13.7394* | 1.3470 | .000 | -17.424 | -10.055 | ||
| GRADUATE | LT HIGH SCHOOL | 46.2190* | 1.2495 | .000 | 42.799 | 49.639 | |
| HIGH SCHOOL | 35.5267* | 1.0936 | .000 | 32.531 | 38.523 | ||
| JUNIOR COLLEGE | 26.2311* | 1.7279 | .000 | 21.492 | 30.970 | ||
| BACHELOR | 13.7394* | 1.3470 | .000 | 10.055 | 17.424 | ||
| *. The mean difference is significant at the 0.05 level. Questions and Answers relating to the Tables Above |
- What is your research question? How can we use socioeconomic index of 2010 to determine whether means differences exist in respondents’ highest degree level?
- What is the null hypothesis for your question? Looking at the socioeconomic index of 2010, there seemed to conclude little to no differences in respondents’ highest degree level.
- What research design would align with this question? The research design I used is the comparative design of methodology. In social sciences, this type of research design is aimed at defining correlation or differences among multiple variables (Cantrell, 2011).
- What dependent variable was used and how is it measured? The dependent variable is R’s socioeconomic index of 2010. It is measured as interval/ratio variable by using the One-Way Anova as a comparison of means test.
- What independent variable is used and how is it measured? The independent variable is R’s highest degrees level. Using the One-Way Anova, it is measured as nominal variable to compare means test.
- If you found significance, what is the strength of the effect? Using Post Hoc Tests and the options for levene statistics, we can conclude the strength of statistical significance to be P<0.05=0.000.
- Explain your results for a lay audience and further explain what the answer is to your research question. Unlike independent, paired and one-way sample t-test, ANOVA involves multiple expectations concerning the system of sampling, the level of measurement, the structure of the sample distribution and the homogeneity of variance (Wagner, 2016). The F-statistics is an important value that helps in determining the significance of our test. Looking at the table, we can see that the significance level is 0.000. This level is far under the conventional threshold of 0.05. Therefore, we can answer our research question by rejecting the null hypothesis that there are no differences in socioeconomic index of 2010 across respondents’ highest degree level. It is essential to find other possible differences and where or how they relates. We can find which means differ using the Post-hoc test. Using Post-hoc test, I have selected the Bonferroni Test for equal variances assumed, and Games-Howell test for equal variances not assumed. I have likewise used the levene’s test or test of homogeneity of variances to determine null hypothesis. In the table, if I look at the significance level, I will see that we are at 0.000, which is under the assigned level of 0.05. Therefore, we will reject null hypothesis that variances are equal. The other two options of Post-hoc test (i.e. Bonferroni Test and Games-Howell) proved the same conclusion, and displayed differences in means test among R’s highest degree level.
Reference
Wagner, W. E. (2016). Using IBM® SPSS® statistics for research methods and social science statistics (6th ed.). Thousand Oaks, CA: Sage Publications.
Cantrell, M. A. (2011). Demystifying the research process: Understanding a descriptive comparative research design. Pediatric Nursing, 37(4), 188-9.
