Course talk:PSYC305/2013ST2/ClassProject/3.3.3 Method - Procedure - Data Analysis
- [View source↑]
- [History↑]
Contents
Thread title | Replies | Last modified |
---|---|---|
Data analysis | 7 | 05:39, 6 August 2013 |
Looking good | 4 | 05:28, 6 August 2013 |
Anyone has any idea about how to perform the data analysis? A few things which I think we should do: 1)Code the response (i.e. strongly dislike = 1, strongly like = 7, etc) 2)Calculate the "d" value between male and female for the response to each question, eliminate the questions that show no meaningful gender differences. 3)Use factor analysis (I don't know enough statistics to be able to do it) to further narrow down our scale to fewer meaningful items. ... 4)Finalize the questionnaire
A few questions remain: 1)I don't know what's the point of knowing the big five personality and how we should incorporate it into our results. 2)How is the reliability of our questions going to be assessed?
Hi there, you are definitely on the right track! I'll let others think about and try to reply to your first question. Your second question is a very good point! I think we should definitely report an estimate for the internal consistency reliability of the final scale. I would suggest we use Cronbach's alpha reliability coefficient. Would anyone like to have a go at calculating this from the raw data? If not, Lauren or I could do it.
Hi everyone,
Apparently, no one has calculated Cronbach's alpha yet. I will finish it this afternoon.
How did you go with this? You might be able to find an online calculator or spreadsheet template for this. Or if you have access to statistical software (such as SPSS) you can do it. Maybe too late now!
Hi there, to answer your first question, based on the information I got from results and discussion sections of Big Five Traits, it seems that our aim is to calculate the correlation between gender diagnosticity scores and the Big Five Traits. For example, what is the correlation between masculinity/femininity and each of the Big 5 personality (i.e., Extraversion, Agreeableness, Conscientiousness, Neuroticism and Openness). To incorporate it into the results, only values of p<0.05 are considered significant. Positive r means the trait goes along with the gender diagnocity score, vice versa. R close to 0 also indicate no significant correlation. Hope this will help~
By keeping your first sentence as a great starting point, I have added all the above-mentioned info of Big Five and age analysis to the last paragraph.
The d value description is great - you guys could do the same thing with factor analysis.
I was about to continue writing for factor analysis, but I got confused with the factor loading we did in the result part. What exactly is factor loading? Thanks!
I'll give someone else the opportunity to answer that one!
I read your discussion post in result part and I found it in the textbook! Thanks!
For those who wants to know the answer to the above question: According to the textbook (p.65): "Factor loadings are the indexes of how much of the variation in a personality adjective (eg. aggressiveness) is "explained" by the factor (eg. Extraversion). Factor loadings indicate the degree to which the item correlates with, or "loads on," the underlying factor."