Science Buddies: "Ask an Expert"

Hello,

My research has on one side one concept defined by 6 attributes (this concept is my dependent variable).
Next, on another side I have three parameters (independent variables).

My research goal is to find potential correlations between three independent variables and dependent variable. But the problem I have is the following. SPSS requires a numeric value for dependent variable (i.e. roughly speaking one column), whereas I obtained six numeric values for every of six attributes that constitute the concept (i.e. six columns). Now I can't figure out how to shape my dependent variable so that I could apply mutliple regression in SPSS... So for independent variables I obtained three columns of data, which is ok. My dependent variable cannot be measured once like for example IQ level, which gives one value = one column of data.

Hello ghost291,

Multiple regression analysis is a very complex statistical tool. There are various conditions and assumptions that must be made before using it. From what I know, your independent variables need to be tested on each dependent variable seperately. SPSS allows you to have several independent variables and it will determine the correlation of each of those on the dependent variable. Maybe try doing several multiple regression analyis. You can test your 3 independent variables to 1 dependent variable attribute. See the correlation. Then repeat the process with a different dependent variable attribute. Do this for all of the dependent variable attributes. This will allow you to know which independent variable affected which dependent variable the more.


This website has quite some information on using multiple regression analysis with SPSS.
https://statistics.laerd.com/spss-tutor ... istics.php

Thank you very much for detailed and clear explanation asyed1229 ! It is really helpful and makes sense. Overall, with several multiple regression analysis I would be able to conclude which independent variable has the most influence on the entire dependent concept itself. Since every dependent attribute forms the concept.

I read all the info here https://statistics.laerd.com/spss-tutor ... istics.php
And it looks I need ordinal regression analysis since:

Assumption #1: Your dependent variable should be measured at the ordinal level. (this is the case for me, every attribute was measured using ordinal scale)

Assumption #2: One or more independent variables that are continuous, ordinal or categorical (including dichotomous variables). However, ordinal independent variables must be treated as being either continuous or categorical. They cannot be treated as ordinal variables when running an ordinal logistic regression in SPSS Statistics;

in this case, considering Assumption #2, my independent variables were measured using ordinal scale (Likert scale to be more precise). It seems like I cannot use ordinal regression analysis since my independent variables are not ordinal

And in the description of multiple regression analysis it is written that
If your dependent variable was measured on an ordinal scale, you will need to carry out ordinal regression rather than multiple regression.

Basic requirements of ordinal logistic regression

In order to run an ordinal logistic regression (really looks suitable for my research), there are four assumptions that need to be considered. The first two assumptions relate to your choice of study design and the measurements you chose to make, whilst the second two assumptions relate to the characteristics of the data that you actually collected (i.e., how your data fits the model). These assumptions are:

Assumption #1: You have one dependent variable that is measured at the ordinal level (that's perfectly fine, I got this for dependent variables). Examples of ordinal variables include Likert items (e.g., a 7-point scale from "strongly agree" through to "strongly disagree"), physical activity level (e.g., 4 groups: sedentary, low, moderate and high), customer liking a product (ranging from "Not very much", to "It is OK", to "Yes, a lot"), and so forth.
Assumption #2: You have one or more independent variables that are continuous, ordinal or categorical (including dichotomous variables). However, ordinal independent variables must be treated as being either continuous or categorical. They cannot be treated as ordinal variables when running an ordinal logistic regression in SPSS;

About Assumption #2, I have ordinal independent variables. So I assume I need to consider them as categorical in SPSS (so it would be a category "strongly agree" or "neutral" for example) since as it is written I cannot use ordinal independent variables. HoweverCategorical variables can be further categorized as either nominal, ordinal or dichotomous. So I don't really get can I use independent ordinal variables ?

Hello,

There might be other valid approaches besides ordinal logistic regression. Can you provide more information on the study design and hypothesis?
Also, could you specify the data values for each of the variables? For example
y1: 1, 2
y2: 1,2,3,4
x1: 1,2,3,4
etc. Because it's hard for me to tell from the screenshot.

~Heidi