1

Our research is entitled "Relation of Family Perfectionism on the Career Indecisions for Grade 10, 11 and 12 Students"

Independent variable: grade level

Dependent variable: family perfectionism and career indecisions - both can be measured by survey and using likert's scale to quantify

A sample of the survey can be found in a similar research: http://www.sciencedirect.com/science/article/pii/S1877042810016356

Our main problem is about writing chapter 3(especially on the part about research procedure and statistical treatment), because of the following assumptions that must be followed if we were to use a one-way ANOVA test:

  1. The dependent variable should be approximately normally distributed. This can be checked using Shapiro-Wilk test of normality. If it fails, then we have to use a Kruskal-Wallis H test instead.(Note: We only have a sample size of 20 for each population, so this assumption can't be simply assumed to be true.)

  2. There must be a homogeneity of variances. This can be checked using Levene's test for homogeneity of variances. If it fails, then we have use a Welch ANOVA test instead.

If both assumptions fail, then a transformation will be done.

Can you write all of this "possible cases" in chapter 3(research methodology)? I asked both my research teacher and math teacher(also a statistician), and they weren't able to give me definite answers about this. Any help? This is just the first part of our research, I'm also unsure which post hoc test to use after.

Anonymous
  • 31
  • 2
  • See if this helps you: https://stats.stackexchange.com/questions/97098/practically-speaking-how-do-people-handle-anova-when-the-data-doesnt-quite-mee?rq=1 – kjetil b halvorsen Aug 07 '17 at 14:50
  • Are you wondering about the actual writing part (as in whether you should write this in your methodology) or the statistics part (should you follow those steps to conduct your ANOVA)? – Leonie Evian Aug 07 '17 at 15:02
  • @LeonieEvian yes that's my first problem, I am unsure of how to write it down. Also, my teachers are unable to give me a definite answer(like saying that I should use paired t-test instead, ignore assumption 1, etc.). But what they are saying is making the research reputable. Thanks in advance. – Anonymous Aug 08 '17 at 09:04

0 Answers0