Chapter 3: Statistical Significance Testing
1. The Alpha level decides what ranges of values are used for the test statistic. (t ratio or z)
In general the smaller the alpha level, the smaller the chance of committing a type I error.
2. Statistical power
Nate Burrow
Psych440
Dr. Volk
10/14/2012
Homework: Chapter 9
1.
a.
b.
c.
d.
e.
b = .359, b0=47.35 - Y=47.35+.36X
For Women
B = .2854, b0=44.31 - Y=44.31+.2854X
Y=44.31+.2854(72) - 64.9 in.
Zy=.32 (Zx) Z scores have a mean of 0, therefore the standardized
Nate Burrow
Psych440
Dr. Volk
10/14/2012
Homework: Chapter 9
1.
a.
b.
c.
d.
e.
b = .359, b0=47.35 - Y=47.35+.36X
For Women
B = .2854, b0=44.31 - Y=44.31+.2854X
Y=44.31+.2854(72) - 64.9 in.
Zy=.32 (Zx) Z scores have a mean of 0, therefore the standardized
Nate Burrow
Psych 440
Dr. Volk
9/16/2012
Homework 5
1.
a. 93
b. In the journal they assumed that the two groups had different variances. Whereas
when I calculated the degrees of freedom I assumed they had the same variance.
2. A. it usually makes t larger
1. The goals of data screening are:
a. Identification and correction of data errors
b. Detection and decisions about outliers
c. Evaluation of patterns of missing data and decisions with respect to how to deal
with missing data
2. The researcher needs to
Nate Burrow
Psych 440
Dr. Volk
8/29/2012
Homework #2
1.
a. M = 2
(X-M)
0
1
1
1
2
7
-2
-1
-1
-1
0
5
SUM = 0
(X-M)2
4
1
1
1
0
25
SUM = 32
S2 = 6.4
SD = 2.53
c. No, because the sum of the deviations should always equal zero if the mean is correct.
f. You div
Homework #1
Psyc440
8/23/2012
A sample is representative if it has characteristics similar to those of the population
1.1 Convenience samples are more commonly reported in journal articles.
1.2 Random sampling is more likely to be representative of clearl
Chapter 6
1-14
1.
a. Stess symptoms purported to be statistically significant with a critical F value of
2.68. Since 3.827 is larger than 2.68, the results are concluded to be statistically
significant. However, depression was not statistically significan
Chapter 4
QUALITY CONTROL FOR DATA COLLECTION
Most methods of data collection are susceptible to recording errors or artifacts, and researchers
need to know what kinds of errors are likely to occur.
Researchers who use self-report data to do research on p