SPSS CLASS REVIEW

Import that data

100

What extension do native SPSS datafiles have?

Native SPSS data files have an ".sav" extension

200

CSV and TXT are delimited files. What is a delimiter?

A delimiter is a character used to specify the boundary between separate columns of data. Ex. a CSV file has several columns of data, each separated by a comma (comma-separated values)

300

The two main windows in SPSS us to look at the data are ____ view and ____ view

(1) data view

(2) variable view

400

In this tab, rows represent variables and columns represent characteristics of variables.

Variable View

500

I administer a survey in May and administer the same survey again in September. The data from these surveys need to be merged into one database. In order to merge, should I:

(A) add cases

(B) add variables

(B) add variables

Clean that data

100

What is the purpose of the "label" column in the variable view

The "label" column is where researchers can write notes to themselves about what the variable is. 

200

True or false. Multiple variables in the same dataset can have the same name, as long as they have different labels

False

300

Use which command to create new variables using algebraic formulas (i.e. BMI = (weight)/(height)²)

COMPUTE

400

Which menu item contains the split file and select cases command?

A. Graph menu

B. Data menu

C. Transform menu

D. Analyze menu

B. Data menu

500

My codebook indicates that the value 9 = N/A. What is the best way to tell SPSS to treat this value as missing?

Code it as a discrete missing value in the variable view

Create those vars

100

True or False. When creating a variable from the values of another variable, it is always better to use the "Recode into same variable" option. "Recode into the different variable" is clunky and uses up space.

False! Always preserve raw data

200

I should use this to make sure my collaborators can compute the same variables as me in the same exact way

Syntax!

300

True or false. I am collapsing categories of married (1. single, 2. married, 3. divorced, 4. separated, 5. widowed, etc) into a binary variable married_b (1 = married, 0 = all other). If the participant is missing data for the original married variable, they should be coded as 0 for married_b. 

False. They should be coded as missing

400

I used the drop down menu to compute a new vairable, BMI, in my data. I clicked "paste" to store the syntax needed to compute this variable. But now I don't see my new variable, BMI, anywhere! What happened?

Paste function only pastes the code, it does not execute the code. You need to go to your syntax file, highlight the code you would like to execute and click the green arrow (run) button. 

500

What s the purpose of this syntax:

recode num1 (lowest thru 60 = 1) (85 thru highest = sysmis) into y1.

Recoding values of num1 to create new variable, y1. 

Name that test

100

Is the average weight of monkeys different than 20lbs?

One-sample t-test

200

Is the average age of patients admitted in the morning significantly different than the average age of patients admitted in the afternoon?

Independent samples t-test

300

Did the average number of student absences change after the midterm?

Paired sample t-test

400

I may want to use a nonparametric test when my sample size is small and _____________

I expect (or do not know) that the distribution of the underlying population is not normal

500

When should I check the output for Levine's test? What does it tell me?

When I am running an independent samples t-test. The null hypothesis of Levine's test is that the variances are equal across groups. These findings tell me whether or not I can assume equal variances. 

Interpret that test

100

Pick one: I conducted a two sample t-test to evaluate whether students in SPSS class get better grades than students in SAS class. Which plot is most useful in visualizing the results:

(1) pie chart

(2) bar graph stratified by SAS/SPSS

(3) histogram 

(4) box plot stratified by SAS/SPSS

(4) box plot stratified by SAS/SPSS

200

When conducting a one-sample t-test where we are comparing the average class grade to 90%, we only need to report a p-value.

False. Must provide descriptive information as well

300

Interpret findings for this question: did the average number of absences change after the midterm?

average diff in absences _{(post midterm - pre midterm)} = 1.5, 95%CI (-0.05, 2.0); p-value = 0.21

On average, there were more absences after the midterm. At the level of significance 5%, we fail to reject the null hypothesis. We do not have evidence to suggest that the difference (1.5) is statistically significant. 

400

True or false. 

A significant test result (P < 0.05) means that the

test hypothesis is false or should be rejected.

False.  The P value is the probability of obtaining data as extreme, or more extreme, than those observed if the null hypothesis is correct. 

500

True or false. The interpretation of the 95% confidence interval is that "if repeated many times, 95% of the time the 95%CI will include the true effect estimate"

True