graphs for "relationship between DV and IV"

I was under the impression that you only have to look at one graph (the same scatterplot that you check homoscedasticity and Independence of Error Assumptions) in order to check for the assumption of the correct relationship between the DV and each IV. However, as I look at the power points from class, it says that we ares uppposed to plot the residuals on tY axis and the predicted values on the X axis, and that we are also supposed to plot the residuals on the y axis and EACH IV on the X axis. Does this mean that we need to show ALL of the graphs in our final project assignment?

Final Project - Results

Hi Guys,

Are we supposed to write up our transformed results for the results section, or our original data? I'm not sure.... thanks!

Assumptions (cont.)

Normality of Residuals: you can check this by clicking the "Plots" button in your regression window and checking the two boxes (Histogram and Normal Probability Plot) in "Standardized Residual Plots."

Independence of Residuals (cont.): the same residual vs. predicted plot that you used for "homoscedasticity" and "correct relationship between DV and IV" can be a rough way of testing this. But another (I think better) way would be to make a boxplot similar to Figure 4.4.6 (C) on p. 135 of the text.

To do that:

1) create a residual variable.

a) in your linear regression window, click "Save"

b) check the box for "Standardized" under "Residuals."

c) run the analysis

2) go to Graph --> Chart Builder --> Boxplot --> Clustered Boxplot.

Now you can put the "Standardized Residual" on the Y axis, a categorical variable in the X axis clustered by another categorical variable. The medians should be similar if there is independence of residuals.

Technically, you could also transform height into a categorical variable to cluster around, but you guys are probably getting exhausted already. So don't worry about it.

You wouldn't have to worry about serial dependency because this is cross-sectional data.

Please ask if you have more questions.

Good luck!

Assumptions

Hey guys,

Here are some responses to questions we've received about the assumptions you should be testing for:

• All relevant IVs are included in the model: this is based on theory, so you will not have a statistical output to "prove" that this assumption has not been violated. However, you should be able to explain that the relevant IVs (that pertain to your hypothesis and what you are testing/researching) are included and measured well.

• No measurement errors in the IV's: Again, this is done prior to the data collection. You want to make sure that the measures used to assess the IVs are reliable and valid.

• NOTE: Dr. Kim has not asked you to write up anything about the two assumptions above on your final project. However, they are important to consider in all future research.

• The relationships between IVs and DV are correctly specified: See the comments provided by David on Diana's questions about the final project (Thurs, June 2). You will be able to see a clear violation if one exists for this project.

• Independence of errors: check this with the residual and predicted plot….and Dr. Kim said “Basically, you don't want to see any specific pattern among the residuals that should be randomly distributed around the zero line.”

• Homoscedasticity: Create a scatterplot with the ZPRED and ZRESID and see if there is a pattern of systematic error. See the notes and powerpoints.

Assumptions

How do I create a graph to check for the assumption of Independence of Errors?

Question on order of IVs

In this final project we are unsure of what doctor kim is looking for with which variable to be entered first. Smoking is the IV we are looking at so it should be entered first, but it also states that the other variables (gender, height etc.) are accounted for. Does this mean that we should enter the variables accounted for first or the most important (smoking). Also, I am enetering the height and age in the same block. Is this what is supposed to happen?

Final Project

Do we need to attach the SPSS output for the final project?

Also, what graph do we look at for assumption 3 (relationship between IVs and DV)?