Posted by
Maguin, Eugene on
Apr 24, 2013; 2:18pm
URL: http://spssx-discussion.165.s1.nabble.com/Do-I-need-to-use-a-regression-analysis-tp5719677p5719692.html
Some of your description is oddly phrased for English, so I'd like to make sure I understand what you have.
a) Are both HR and Process categorical variables with three categories?
b) What does '... all variables also have "control variables".' Mean?
c) are the data at the organizational level (so, one record for each organization) or for units or persons within each organization (so, multiple records for each organization)?
d) You say, '... organizations can use both HR and Process, or just one of those. Not all organizations perform Innovation (but some do have HR or Process without Innovation)'. Now: 1) If an organization does not perform Innovation or do not use HR or Process, are there missing data for those variables? 2) If organizations do not perform innovation or do not use HR or Process have missing data, how do you propose to analyze such a data set since the cases used in the H1 analysis were be somewhat different from the cases in the H2 analysis?
Gene Maguin
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of Kim S.
Sent: Wednesday, April 24, 2013 7:15 AM
To:
[hidden email]
Subject: Do I need to use a regression analysis?
Hi Guys,
I have a question regarding my master thesis. In which I have 3 hypotheses.
H1= HR has a positive influence on Innovation In which HR is a categorical variable existing out of low/medium/high Innovation is a ratio/interval existing out of a % of turnover
H2=Process has a positive influence on Innovation In which Process is a categorical variable existing out of low/medium/high Innovation is a ratio/interval existing out of a % of turnover
H3= HR has a greater positive influence than Process on Innovation
The dataset exist out of 400 organizations, organizations can use both HR and Process, or just one of those. Not all organizations perform Innovation (but some do have HR or Process without Innovation)
I don't know yet if the data is normally distributed. And all variables also have "control variables".
In the book of Field i think I have to use Factorial Mixed ANOVA, according to my lectures notes i have to use a Multiple regression analysis and according to my mentor it could also be a partial square modelling? So now I am confused what to do and how to explain why I made this decision.
Hope to hear some from you, as it will help me out lots!
Best regards,
Kim
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Do-I-need-to-use-a-regression-analysis-tp5719677.htmlSent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD