Let’s say you were talking with a friend and they said that males are more risk-taking than females. This makes some sense given what we know about the male species. However, you are a scientist and you say to your friend, “let’s actually see if we can answer this question scientifically rather than based on the stupid things we see guys do compared to the typical calculating approach that females usually take.” In order to do this, you walk through NHST process:
Using the class_survey dataset complete the following:
Part 1: Are there differences in levels of sensation-seeking between males and females?
Come up with a null and alternative hypothesis regarding the association between sex/gender and sensation seeking (Brief Sensation Seeking Scale; BSSS_Total).
First you are to compute descriiptive statistics (i.e., mean, min – max, standard error of mean, and standard deviation) of sensation-seeking for males and females separately. To do this you will have to use the “select cases” feature in spss.
select “data”
from the drop-down menu select “select cases”
in the dialogue box under “select” click the radio button next to “if condition is satisfied” then click “if”.
find the gender variable in the left-hand column and move it over to the open window on the right using the arrow button.
to select on males put an “=” sign next to gender
and set it equal to “0”.
click continue and then OK.
to check and make sure it worked you can run a frequency on gender and only males should be reported. You will also see that in data view the female cases will have a slash through them.’
compute the mean of BSSS_Total.
follow the same steps for females (i.e., select cases = 1 for gender).
Note that when you do this you will see a new variable created “filter_$”. This is so you can easily apply the filter later on but you do not need to keep this variable so do not save the dataset when closing the application (or just delete the variable).Based on your output, are there mean differences between males and females on sensation seeking? Describe this in a couple of sentences. You should discuss the direction and how much of a mean difference there is. Keep in mind that by just observing the means, we cannot say any difference is significant. We can only say that the means are different in our sample – we cannot draw any inferences about he population.
The next thing you are going to do is compute 95% confidence intervals for males and females by hand to see if they come from the same population regarding sensation seeking (see section 2.8: “I is for confidence interval” in your text book). You will need to use the appropriate information from the output produced in #2 (mean and SE of the mean). You may also need to use the standard normal table. You also need to show your work for computing the 95% C.I. When doing this you can round out to the third decimal place. Please type this out in the Word document.
Now you are to check that you did this properly by computing 95% confidence intervals in spss (see video in module). Note. Make sure that you go back to “select cases” and click the radio button for “all cases”. The answers you obtained when computing by hand may not be exactly as what spss gives you due to rounding but they should be pretty close. The output from this is to be copied and pasted into a word document with the rest of your work.Discuss what you observed regarding sensation seeking and sex in paragraph form and whether or not you should “reject” or “fail to reject” the null hypothesis based on your findings. In other words, determine and discuss if 95% C.I.s overlap (or not) and what that means regarding your hypothesis.
Part 2: Are there differences in sensation-seeking between those who have engaged in theft and those who have not engaged in theft?
Repeat steps 1 – 6 of Part 1 except compare those who have reported intentionally stealing something to those who have not on sensation seeking. You will use the “Theft” variable and the “BSSS_Total” variable.
Your completed assignment (turned in to me) should include: your two sets of hypotheses, write-up for #3 (for Parts 1 and 2), results for 95% C.I.s (show work, for Parts 1 and 2), copied and pasted output for the 95% C.I. from spss, write-up regarding whether or not you found support for the hypotheses (#6; for Parts 1 and 2). All should be in a word document with your first and last name in the file name.