PSY 3211 – Research Methods and Design I – Discussion #13 – Three’s A Crowd
Group Members: _______________________________________________________________
Instructions: FIU has a pretty good reputation for addressing the needs of traditional students
(those who attend classes on campus), fully online students (who attend classes only online), and
hybrid students (those who have both online and on campus course components). Let’s say we
want to test the effectiveness of these different learning modalities when it comes to a class in
psychology research methods. You design a study to assess methods knowledge at the end of the
semester (scores can range from 0 to 100 on a 100 point scale). Using the data below for 10
students in each condition, calculate the F value and determine if it is significant. Make sure to
complete the eight steps for research noted in your Salkind textbook. Finally, write up the
findings as you would see it in an APA empirical journal article (if significant, you would
normally need to run a post hoc test. However, you can just provide the means and eyeball them
to see which, if any, learning modality leads to more learning).
As you work on this problem, go through the each of the eight steps for research noted in your
Salkind textbook again. This is a tough assignment, so I will give you the means and SDs for the
first two (find the hybrid mean and SD on your own):
Class
Mean
Standard Deviation
Traditional Scores
96
94
93
95
86
80
84
77
86
92
88.30
6.65
Online Scores
93
78
91
67
85
99
88
82
92
84
85.90
9.00
Hybrid Scores
96
77
89
80
93
91
94
92
97
82
Use the last page
Use the last page
As you work on this problem, make sure to provide information for each of the eight steps we
cover in Chapter 11 (Salkind) as well as the APA write-up you would see in a results section.
1. State the null and alternative hypotheses
2. Tell me your level of risk
3. Determine the best statistical test to use
4. Compute the test statistic. I’ll even give you the table to use (formulas on the last page)!
Tradition
96
94
93
95
86
80
84
77
86
92
X2
Online
93
78
91
67
85
99
88
82
92
84
X2
Hybrid
96
77
89
80
93
91
94
92
97
82
n
ΣX
X
Σ(X2)
(ΣX)2/n
X2
N
ΣΣX
(ΣΣX)2/N
ΣΣ(X2)
Σ(ΣX)2/n
SS Between =
SS Within =
Total Sum of Squares =
MS (between) =
MS (within) =
F = MS Between / MS Within =
5. Determine the value needed to reject the null hypothesis. Remember to calculate the
correct degrees of freedom before finding the critical F-value!
6. Compare the obtained and critical value
7. Decide whether you will retain the null hypothesis or …
8. Decide whether you will reject the null hypothesis
9. Write up your results as you would see it in a results section of an empirical research
paper. Make sure to include the means and SDs for all three conditions. If significant,
note that a post hoc test would be needed to see which means differ, but you do not have
to run actual post hoc tests for this discussion.
10. Was your obtained F value in #4 positive or negative? Would it matter either way? With
your discussion group, tell my why a positive or negative value is not important when it
comes to your obtained value. (Note: This question is a bit trickier to answer than the
similar one I asked in Assignment #11. There, we discussed the t-Test and
positive/negative values. The answer to this one in Assignment #13 is different, but you
should still be able to answer it!)
11. We saw in Chapter 12 (Tea for Two Again) that we can run a dependent t-Test. Can we
do the same (that is, use the same participants multiple times) with a three group design?
If so, design a dependent study with three groups. Make sure to note your IV and DV.
12. Although we won’t cover post hoc tests in this course, we will in Methods II. I want to
prepare you for them a bit right now. Look at figure 13.6 in your Salkind
textbook(pp.258). With your group, tell me which of their three conditions differ (if any)
in your textbook example and tell me how you know they differ.
13. When should you run post hoc tests? Why do you run them? Finally, why not just run
three different t-Tests, one comparing groups 1 and 2, one comparing groups 1 and 3, and
one comparing groups 2 and 3? Justify your responses.
MSbetween =
SSbetween
dfbetween
SSbetween =
MS within =
( x) 2 ( x) 2
−
, dfbetween = k − 1
n
N
SS within
df within
SS within = ( x 2 ) −
( x) 2
, df within = N − k
n
Here’s a table you can use for the SD calculation! As promised, I did the first two groups.
Traditional
96
94
93
95
86
80
84
77
86
92
M=
n–1
(X – M)2 /
n–1
SD (√ )
X–M
7.7
5.7
4.7
6.7
-2.3
-8.3
-4.3
-11.3
-2.3
3.7
(X – M)2
59.29
32.49
22.09
44.89
5.29
68.89
18.49
127.69
5.29
13.69
Online X – M
93
7.1
78
-7.9
91
5.1
67
-18.9
85
-.9
99
13.1
88
2.1
82
-3.9
92
6.1
84
-1.9
(X – M)2 Hybrid X – M (X – M)2
50.41
96
62.41
77
26.01
89
357.21
80
.81
93
171.61
91
4.41
94
15.21
92
37.21
97
3.61
82
88.30
9
398.1 / (10 – 1) = 44.23
85.90
9
728.9 / (10 – 1) = 80.99
6.65
9.00