Discussion Forum #5: Proposal Stats
Part #1.) Main Post
DF5: Proposal stats
This discussion forum should help get you started on the deciding on your data analyses for your study, which will be needed when you have to submit your method section draft very soon.
So, for this post I want you to do the following:
I intend to use independent samples t-test to find out whether there will be significant differences in change scores between the experimental and the control groups. I believe that there is a causal-effect relationship between the use of the internet social networking tools like Facebook and communication behaviors and social interaction skills. An Independent samples t-test will make it possible to determine if use of Facebook has any effect on communication behaviors and social interaction skills. The test is suitable for the study because there will be two unrelated group.
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
What will this analysis tell you? For example, if you have a significant correlation coefficient, what does that mean for your variables. (2 points)
The size of the p-value, which is associated with the F-statistic, will tell whether the difference is statistically significant or not. If the value of p is below the conventional significance level of .05, it means that the difference is statistically significant. Therefore, I will conclude that use of social networking tools like Facebook has a reliable effect on communication behaviors and social interaction skills.
_______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
C. Provide feedback to your classmates (at least two) regarding their analysis chosen (2 points total)
You should post your initial post by Sunday night and add your feedback to a classmate by Monday night. You only get all 10 points if you give meaning/helpful feedback to your classmates by the deadline of Monday.
Part C#1.) Reply Post
df5
Sunday, 21 February 2016, 11:16 PM
I plan on using the Pearson r correlation coefficient analysis to determine the strength of the relationship between physical fitness and mood levels in the participants._______________________________________
__
Feedback
_
It will be possible to determine the strength of the correlation between physical fitness and mood levels. I also wish to add that it will be useful to test the data to see whether it meets the assumption of normal distribution. It will be possible to do this using the Shapiro-Wilk test of normality which is available in SPSS.
_____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
The sign of the Pearson r tells us about the direction of the relationship; that is, whether there is a positive relationship or a negative relationship between the variables (Cozby Bates 2015). If the correlation is strong, it will support my hypothesis.
Feedback
As you say, the sign will give the direction, either positive or negative. I would like to add that correlation coefficients that will be close to 1 regardless of the sign before it will be the strongest. Those that will be close to 0 will be the weakest.
________________________________
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
Part C#2.) Reply Post
DF5
Wednesday, 17 February 2016, 7:41 PM
I would like to do a Pearson r correlation coefficient statistical analysis for my study, because I predict a linear relationship between my variables. My hypotheses focus on two parenting styles (authoritative and authoritarian), and their relationships with self-esteem, mood, and grade. My x variables would be the PAQ scale scores. My y variables would be the GPA's, self-esteem scale scores, and the mood scale scores.____________________________________________
Feedback
The use of Pearson r correlation coefficient is suited for this analysis. I just wish to add that it will be good if you remember to see whether the most important assumptions such as approximately normal distribution are not violated. You will be able to do this by using the Shapiro-Wilk test of normality available in SPSS.
____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
When testing the relationship between authoritarian parenting and self-esteem, I hope to see a negative correlation that at least represents a strong relationship. (a negative r-value that ranges between -.4 to -.99).
When testing the relationship between authoritarian parenting and mood, I hope to see a negative correlation that at least represents a strong relationship. (a negative r-value that ranges between -.4 to -.99).
When testing the relationship between authoritarian parenting and grades, I hope to see a positive correlation that at least represents a strong relationship. (a positive r-value that ranges between .4 to .99).
When testing the relationship between authoritative parenting and self-esteem, I hope to see a positive correlation that at least represents a strong relationship. (a positive r-value that ranges between .4 to .99).
When testing the relationship between authoritative parenting and mood, I hope to see a positive correlation that at least represents a strong relationship. (a positive r-value that ranges between .4 to .99).
Feedback
I agree with you that correlations close to 1 such as .99 are the strongest; and that if they are obtained, then there will be a strong relationship. I would also wish to add that you should not forget to determine whether the coefficients are statistically significant. To determine whether they are significance, you will need to look at the p-values. If you will have set your significance level at .05, anything below it is statistically significant and anything above it is not statistically significant, meaning the correlations might have been as a result of chance.
___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
