Regression and Adjusted Means

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Kingscrown21
Posts: 2
Joined: Sat Oct 12, 2013 2:33 am

Regression and Adjusted Means

Postby Kingscrown21 » Sat Oct 12, 2013 4:54 am

I’m trying to replicate a couple of tables from an article for a project. I was able to replicate the first table in which the author was looking at Mean Church Attendance broken down by Sexual Orientation. (I did this using compare means with Church Attendance as the Dependent Variable and Orientation as the Independent Variable)

The second table is similar but is looking for Adjusted Mean Church Attendance broken down by Orientation and is Calculated from an OLS Regression Model. The Dependent Variable again is Church Attendance with Control Variables being: Age, Education, Race, Marital Status, No Religious Affiliation, and Number of Children.

I have all the appropriate variables and have recoded them where necessary. What I am unsure of is how do I set up this regression appropriately and how do I use the results of the regression to get the new adjusted mean values for Church Attendance?

Any help is greatly appreciated.
statman
Administrator
Posts: 2757
Joined: Tue Jun 12, 2007 12:08 pm
Location: Florida, USA

Re: Regression and Adjusted Means

Postby statman » Sun Oct 13, 2013 11:48 am

See the Note below
See the note below

NOTE: Please read the Posting Guidelines and always tell us your OS, the SPSS version and information about your study and data!

Statman
Statistical Services
Kingscrown21
Posts: 2
Joined: Sat Oct 12, 2013 2:33 am

Re: Regression and Adjusted Means

Postby Kingscrown21 » Mon Oct 14, 2013 1:21 am

statman wrote:See the Note below
My apologies.

OS Name Microsoft Windows XP Professional
Version 5.1.2600 Service Pack 3 Build 2600

IBM SPSS Statistics 20

I’m doing a replication and update to a previously published study. The study uses data from the General Social Survey (years 1991-2000) and looks at the relationship between gender, sexuality, and religious commitment.

The author uses four different variables to measure religious commitment:

Church Attendance - Likert Scale Ordinal Variable
Frequency of Prayer – Likert Scale Ordinal Variable
Biblical Belief – Nominal Variable
Apostate - Nominal Yes/No Variable created from two other variables (religious affiliation at age 16 and current religious affiliation) which measures if an individual has “left” religion.

Biblical Belief was recoded into a Yes/No Dummy Variable. Church Attendance, Frequency of Prayer and Apostate did not need to be recoded

A Variable for sexual orientation was created combining the variable for Gender and the variable for Sex of Sex Partners. The variable had six categories: Female Heterosexual, Male Heterosexual, Female Bisexual, Male Bisexual, Gay Female, and Gay Male.

The first Table in my study looked at means (for Church Attendance and Frequency of Prayer) and percentages (for Biblical Belief and Apostate) for the 4 religious commitment variables for each of the Sexual Orientation categories.

I got the exact values for this table by performing a compare means in SPSS with the 4 religious commitment variables as dependent variables and Sexual Orientation as the independent variable.

The second table is similar to the first but looks at just Church Attendance and Apostate. For Church Attendance it wants the adjusted mean after performing an OLS Regression controlling for Age, Education, Race, Marital Status, No Religious Affiliation, and Number of Children. For Apostate it wants the adjusted predicted probability after performing Logistic Regression controlling for Age, Education, Race, Marital Status, and Number of Children.

I have all the Control Variables and recoded them where necessary. I need help properly setting up the regressions as well as figuring out how to use the results of the regression to get the adjusted mean and adjusted probability.

Who is online

Users browsing this forum: No registered users and 2 guests

cron