Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

Data Mining in SAS After doing a stepwise feature selection I get the following

ID: 3055080 • Letter: D

Question

Data Mining in SAS

After doing a stepwise feature selection I get the following results:

What I am trying to predict is if a person is a morning person or not.

The dataset is a survey that has the following question

Do you find it hard to get up ealy? 0 = disagree or 1 = Strongly Agree

My question is, How do I interpret the results?

For Puncuality the survey had the following options:

1 = I am often early.

2 = I am always on time.

3 = I am often running late.?

Is there a way to know that people that are morning people are more punctual? I really do not know how to interpret the results. Note: After doing the Logistic regression my model is faily accurate.

Summary of Stepwise Selection Effect Number Score Wald Step Entered Removed DF 1Punctuality 2 Prioritising_workloa 3 Finances 4Energy_levels 5 Cheating in school 6 Spiders 7 Entertainment_spendi 8 Documentary 9 Elections 10 Law 11 Criminal_damage 12 Rock In Chi-Square Chi-SquarePr> ChiSq .0001 .0001 0.0002 0.0015 0.0015 0.0060 0.0108 0.0276 0.0187 0.0144 0.0329 0.0320 34.9373 2 23.3545 13.5493 10.1010 10.0572 7.5358 6.4914 4.8506 5.5257 5.9866 4.5504 4.5963 10 12

Explanation / Answer

As per your question , your response variable is morning or not .

You want a interpret the punctuality variable which is a categorical variable .  

And catagorica are

1-i am early

2- i am always early

3 - i am oftrenearly

When you run the regression it will give the significance p-values for the point 1 , 2 and 3 .

If above there points are significant then there must be on category in base line .

Suppose we put( 1- i am often early ) in a base one category. Then result depend on it .

Suppose the punctuality estimate for (2- i am always on) is 0.4 then , we can conclude that , if effect of the punctuality(2-i am always on) on response variable is e^0.4 as a baseline category is (1-i am often early )

Suggestions- suppose you are hard to understand the logistic regression then simple you can use the dummy stepwise regression . You can easily interpret it .

Thanks in advance