Stat Exercise 1: Descriptive Statistics
(Due April 4)
SPSS Work:
A researcher conducts a survey on Facebook use by college students. Here is the 5-question scale
she used to measure attitudes toward Facebook use:
Please rate how much you agree with each
statement
1. I check Facebook every day.
2. I get annoyed when people I don’t know
very well contact me through Facebook.
3. I would feel lost if I could not get on
Facebook daily.
4. Contacts on Facebook are superficial.
5. Facebook helps me feel close to my
friends.
Strongly
Disagree
Disagree
Neutral
Agree
Strongly
Agree
1
1
2
2
3
3
4
4
5
5
1
2
3
4
5
1
1
2
2
3
3
4
4
5
5
She collected data from 10 people; here is a summary of the data:
Facebook Questionnaire
ID
Age
Gender
1
2
3
4
5
6
7
8
9
10
19
18
23
22
21
45
19
50
23
28
Female
Male
Female
Male
Female
Female
Female
Male
Male
Female
Q1
4
5
4
5
4
1
4
2
5
4
Q2
2
1
2
2
1
3
3
4
2
1
Q3
3
5
4
4
4
1
5
1
4
3
Q4
2
1
2
2
1
5
2
5
2
1
Q5
3
5
5
4
4
1
4
1
4
3
PART A (SPSS data entry and preparation).
Enter the data in SPSS. You can make up whatever variable names make sense to you,
but be sure to include variable and value labels. All variables should be numeric, so be
sure to code gender as a number. (see SPSS guide data creation)
Look over the questionnaire. You will want a higher number to indicate a more positive
attitude toward Facebook use. Right now there are some questions that need to be reverse
coded.
Recode these into different variables in SPSS (When naming the new variable, we
recommend adding an “r” to the original name, recode so that a score of 1=5, 2=4, 3=3,
4=1, 5=1). (spss recoding)
Compute the total score on the Facebook questionnaire by computing the sum of the 5
questions. Remember to use the RECODED variables here, not the original ones. Come
up with a name for your new (summed) variable that makes sense.
Check your data for errors by manually checking your data entry and/or by using SPSS.
You should have no missing data – if you do, you made an error somewhere along the
lines. If there are errors, make corrections to your data set and re-check until you are sure
you have no errors with date entry, recoding, or computing.
**Save your dataset (.sav), but DO NOT print it!! You will use it for a future homework
assignment.**
PART B (Data Analyses).
1. Calculate the frequency of males and females using SPSS. (spss frequencies)
2. Using the Frequency command, calculate the mean, median, min, max, and SD for age.
Request a histogram (including the normal distribution) to represent the distribution.
3. Using the Frequency command, calculate the mean, median, min, max, and SD for the
total attitude score you computed. Do NOT request a histogram.
4. Using the Descriptives command, calculate the mean, min, max, and SD of the total
attitude score for males versus females (use the “Compare Means” command and
select “options”). (spss descriptives)
5. Using the Descriptives command, calculate the mean, min, max, and SD of total
attitude scores for all participants in the sample who are under 25 (use the “select cases”
command). (spss data creation)
6. Compute the internal consistency for the Facebook questionnaire by calculating
Cronbach’s alpha. (spss cronbach’s alpha)
**Save your output file (.spv). Submit it along with your answers to the questions in PART
C **
PART C (Results/Interpretation).
1. Does the histogram of age suggest a normal distribution? If not, describe its shape.
2. What is the best measure of central tendency for each of the following variables? Explain
each answer by describing what scale of measurement is represented, and, for interval or
ratio variables, if the distribution is or is not within acceptable limits for the normal
curve.
Gender
Age
Total Score
3. Write a few sentences describing the sample as you would in a Participants section.
Make sure you use the appropriate statistics, and you must match your SPSS output.
(Hint: You should indicate the number of males and females, the dispersion of age (min/max or range),
and the median age.)
4. Write a few sentences describing the questionnaire as you would in a Measures section.
Comment about your results for internal consistency of the questionnaire. Does it meet
the .70 standard?
(Hint: You should indicate the number of items, the type of scale (response categories), and Cronbach’s
alpha)
5. Write a short paragraph appropriately describing the Results, including a) total attitude
score for the entire sample; b) for men and women separately; c) for those under 25.
Make sure you report the appropriate statistics in the Results, and the results you report
must match you SPSS output.
(Hint: You should be clear and concise, indicate the M & SD for the total score, the M & SD of the total
score for men vs. women, and the M & SD of the total score for those who are under 25.)
Hand Calculation:
1. The number of dogs owned by four families are as follows: 1, 2, 2, 3
a. Calculate the mean, median, mode, range, variance, and standard deviation by hand. For
variance, divide the sum of mean squared deviation by n-1, not n.
b. A fifth family is added to the sample. This family owns 50 dogs. Re-compute the
statistics in a).
c. Which of the statistics are not much affected by the inclusion of an observation that is
far away from the rest?
d. For the family who owns 50 dogs, calculate its standard score and percentile for the
number of dogs owned. You can use the z table (in the “excel demo” folder) to find the
percentile.
2. If grades on an exam are approximately normally distributed, with an average 70 and
standard deviation of 10, what percentage of the students:
a.
received grades less than 70?
b.
received grades greater than 70?
c.
received grades less than 60?
d.
received grades less than 50?
e.
received grades less than 50 or greater than 90?
f.
received grades greater 85?
g.
received grades less than 45?
h.
received grades between 45 and 85?
(for f, g, h, use the z table in the “excel demo” folder in WyoCourses)
Purchase answer to see full
attachment