Explore BrainMass
Share

# smokers and nonsmokers on depression scores

The 4th step in hypothesis testing is to compute your data.

Creswell (2008) gave a list of questions to ask yourself to determine what tests to use:

Do you plan to compare groups or relate variables in your hypothesis or research questions?
How many independent variables do you have in one research question or hypothesis?
How many dependent variables do you have? (typically one but if more than one usually looked at one at a time)
How will your independent variable be measured? (ratio are continuous variables)
How will your dependent variable be measured?
How are your scores distributed? Can you assume a normal curve?

For these null hypothesis what type of measures will you use?

There is no difference between smokers and nonsmokers on depression scores.
There is no difference between smokers and nonsmokers and peer group affiliation.

#### Solution Preview

There is no difference between smokers and nonsmokers on depression scores.

Do you plan to compare groups or relate variables in your hypothesis or research questions?
- In this case, we would randomly select individuals to be part of the study, and we would try to get smokers and non-smokers to join. We would then administer a depression test to each individuals. Please note that we can't get pure randomization here, since we can't randomly assign people to smoke or not smoke. People have self-selected themselves into these group. This is called a ex-post-factos non-experimental design. There are inherit biases associated with this type of experiment since perhaps people who are prone to smoking are more depressed in general - but there is no way to ethically get rid of these biases. All this to say that we plan to relate variables in our research question - score on a depression test.

How many independent variables do you have in one research question or hypothesis?

We have one IV - the degree of smoking. There can only be one IV in simple hypothesis testing. Why? If we have more then 1 IV, then we are never going to be sure of which IV caused the DV. In other words, if we are looking at smoker and non-smokers and wages earned, we will not be sure which variable effected the depression score. Are people more depressed because they smoke, or is it because they earn minimal wages? We need to remove all biases and focus on one IV at a time.

How many dependent variables do ...

#### Solution Summary

Results for smokers and nonsmokers on depression scores are presented.

\$2.19