statistical Analaysis? NEED HELP... URGENT!!!

#1
I am working using secondary data that was originally collected for another purpose. The data is from a teen pregnancy program among Latino youth. The program had a control group and an intervention group and collected baseline and follow up data from both control and intervention. the intervention group of the program focused on teen pregnancy only and the control group was only focused on physical activity and dietary intake.



my thesis is only focusing in control group of the program that was focusing in dietary intake and physical activity. for the purpose of my thesis, the survey questions that were related to dietary intake and physical activity and also socio-demo were used.


so for my research question 1.. I need to find out if there is relationship between socio-demographic variables (gender, language spoken, years in US/US born, mother's education level) and physical activity, fruit and veg intake, fast food consumption and soda comsumption).


these are my socio-dem variables (independent variables):

gender (1=male, 2=female)

language spoken (1=english, 2=spanish)

Years in US/US born (0=US born, 1=less than 1 year, 2=1-5 years, 3=6-10 years, 4=more than 10 years)

mother's education level= (1=less than high school/GED, 2=graduated High school, 3=some college/vocational school, 4=graduated college)


these are my dependent variables:

fruit/veg consumption (1=yes, met the recommendation, 2=No, didn't meet the recommendation)

physical activity (1=yes, met the recommendation, 2=No, didn't meet the recommendation)

fast food consumption (1=less than 1 time per week during the past 7 days, 2=1 or more times per week during the past 7 days)

soda consumption (1=less than 1 time per week during the past 7 days, 2=1 or more times per week during the past 7 days)


for my research question 2: I want to see if there is any change at baseline and follow up for control group using above dependent variables (fruit/veg consumption, physical activity, fast food consumption, soda consumption).
 

Karabiner

TS Contributor
#4
Re: statistical Analaysis?

If you only want to test bivariate relationships, then you
can analyse the associations between your dependent
variables and language & gender using Chi² tests.
You can analyse the associations between your
dependent variables and the ordinal variables (years in
US & edlevel) using Mann-Whitney U-tests (technically,
you will have to use the dependent variables as grouping
factors there).

Regarding your second research question, you can use
McNemar's test, which is appropriate for dichotomous
pre-post data from the same subjects.

With kind regards

K.
 

noetsi

No cake for spunky
#7
I always wondered if posts thought that typing that increases the chance of a question being answered. In practice having a question answered depends largely on whether posters actually work in that area of statistics (and how busy they are at any given moment).
 

Karabiner

TS Contributor
#8
At times, many OPs (often in the psychology forum & the SPSS forum,
which seem overlap to large degree) give their threads 5 stars from the
beginning, but AFAICS the frequency of answers is actually lower for
those postings than for postings without "excellent" self rating. Might
be because the questions are comparingly lousier, and/or people (like
myself) don't like such practices.