Lab 292 Assignment 2 (draft)

advertisement
Running head: PSYCHOLOGY 292 LAB ASSIGNMENT
1
Lab Assignment #2
Madison Links
201102207
presented to Instructor J. Morgen
in Psychology 292 Lab
Psychology Department
St. Francis Xavier University
3/28/2014
LAB ASSIGNMENT # 2
2
1. Hand calculations on separate page.
2. a) The null hypothesis would be that there would be no moon effect, and therefore the diameter of the circles draw by participants
would be the same as the diameter of the moon shown in the image (H0: µ=1). The research hypothesis states that there is a moon
effect, and therefore the diameter of the circles that the participants had drawn would be different than the actual diameter of the moon
in the image (H1: µ≠1).
b) This test requires that a single sample t-test be conducted, because there is only one variable involved in this study and that is the
diameter of the moon drawings done by the participants, and the scores are presented in a ratio compared to the actual diameter of the
moon in the image. The mean of the sample ratios needs to be computed and compared to the ratio when there is no moon effect (ratio
= 1).
c) A single sample t-test was performed and a significant difference was found, t (9) = 4.298, p < .003 (two tailed), therefore reject the
null hypothesis and conclude that there was a significant difference between the diameters of circles drawn by the participants, which
were represented by a ratio that compared values to actual size of the moon in the image, and the ratio that represents the actual
diameter of the moon in the image. The scores gathered from participants (M =1.46, SD=0.34) demonstrates that the participants had
drawn larger diameters of the moon compared to the actual diameter of the moon in the image (µ=1).
Evidence from SPSS:
One-Sample Statistics
N
moonratio
10
Mean
1.4630
Std. Deviation
.34069
Std. Error
Mean
.10773
One-Sample Test
Test Value = 1
95% Confidence Interval
of the Difference
moonratio
t
4.298
df
9
Sig. (2-tailed)
.002
Mean
Difference
.46300
Lower
.2193
Upper
.7067
LAB ASSIGNMENT # 2
3
3. a) The null hypothesis in this situation is that the mean difference between the remaining amount flat crust and regular crust is zero,
and this means that there was no difference in the amount of each type of pizza remaining after 15 minutes (Ho: µ=0). The alternative
hypothesis would be that there is a difference in the amount of pizza remaining because people will eat more of the pizza that he or
she prefers, and therefore the mean difference between the two types of pizza is not zero (Ho: µ≠0).
b) This study requires that a dependent samples t-test be run to analyze the data. This type of test should be run because there are ten
participants all of whom participated in both the flat crust and regular crust conditions of this experiment. In order to compare the
means a dependent samples t-test must be conducted.
c) A dependent samples t-test was performed and no significant difference was found, t (9) = (-2.939), p > 0.01 (two tailed), therefore
fail to reject the null hypothesis,. At the significance level of 0.01 there is no significant difference between the amount of flat crust
and regular crust that remained after 15 minutes.
Evidence from SPSS:
Paired Samples Statistics
Pair 1
FlatCrust
Mean
8.3600
RegularCrust
12.2100
10
Std. Deviation
4.82429
Std. Error
Mean
1.52557
10
3.60569
1.14022
N
Paired Samples Test
Paired Differences
df
Sig. (2-tailed)
99% Confidence Interval
of the Difference
Pair 1
FlatCrust - RegularCrust
Mean
-3.85000
Std. Deviation
4.14253
Std. Error
Mean
1.30998
Upper
-8.10723
Lower
.40723
t
-2.939
Std. Deviation
9
Std.
Error
Mean
.017
LAB ASSIGNMENT # 2
4
Paired Samples Test
Paired Differences
df
Sig. (2-tailed)
95% Confidence Interval
of the Difference
Pair 1
FlatCrust - RegularCrust
Mean
-3.85000
Std. Deviation
4.14253
Std. Error
Mean
1.30998
Upper
-6.81339
Lower
-.88661
t
-2.939
Std. Deviation
9
Std.
Error
Mean
.017
d) If this test were conducted using a less stringent significance level such as 0.05 then the conclusion would be different, because the
results would support the alternative hypothesis. A dependent samples t-test was performed and a significant difference was found,
t (9) = (-2.939), p < .018 (two tailed), therefore reject the null hypothesis and conclude that participants ate significantly more of the
flat crust pizza (M = 8.36, SD =4.82) than the regular crust pizza (M=12.21, SD=3.60).
4. a) The null hypothesis for this study would be that there is no difference in scores between those who listened to the tape while in
the car, which was the experimental condition, and those who listened to the tape in control condition held in the lab (H0: µc= µe). The
alternative hypothesis would be there is a difference in scores between those who listened to the tape while in the car and those who
listened to the tape in the lab (H1: µc≠ µe).
b) An independent samples t-test should be conducted for this study. There are two separate groups of participants, and these include
participants in the experimental condition and the control condition. The data collected from participants in each condition are
independent from one another and therefore the means must be compared using an independent samples t-test.
c) An independent samples t-test was performed and no significant difference was found, t (18) = (0.685), p > 0.05 (two tailed),
therefore fail to reject the null hypothesis and draw no conclusions.
Evidence from SPSS:
LAB ASSIGNMENT # 2
5
Group Statistics
Errors
Condition
control condition
N
experimental condition
10
Mean
10.5000
10
9.7000
Std. Deviation
3.37474
Std. Error
Mean
1.06719
1.49443
.47258
Independent Samples Test
Levene's Test for
Equality of Variances
t-test for Equality of Means
95% Confidence
Interval of the
Difference
F
Errors
Equal variances
assumed
Equal variances
not assumed
Sig.
7.874
t
.012
df
Sig. (2-tailed)
Mean
Difference
Std. Error
Difference
Upper
Lower
.685
18
.502
.80000
1.16714
-1.65208
3.25208
.685
12.399
.506
.80000
1.16714
-1.73394
3.33394
5.a) The probability of guessing each question correctly is 1/5, which is 20%. The null hypothesis is that students will not perform
differently than chance when guessing on multiple choice questions (H0: µ=20%). The research hypothesis states that students are
able to perform differently than chance when guessing on multiple choice questions (H1: µ≠20%).
b) This test requires that a single sample t-test be conducted, because there is only one variable involved in this study and that is the
students’ scores on the test. The sample mean of the scores must be computed and compared to the probability of guessing each
question right (0.2).
c) A single sample t-test was performed and a significant difference was found, t (27) = 20.613, p < 0.001 (two tailed), therefore reject
the null and conclude that students are able to perform differently than chance when guessing on multiple choice questions.
Participants scored significantly higher (M=46.21, SD=6.73) compared to performing purely by chance (µ=20).
Evidence from SPSS:
One-Sample Statistics
N
StuScores
Mean
28
46.2143
Std. Deviation
6.72947
Std. Error Mean
1.27175
LAB ASSIGNMENT # 2
6
One-Sample Test
Test Value = 20
t
df
Sig. (2-tailed)
Mean Difference
95% Confidence Interval of the
Difference
Lower
StuScores
20.613
27
.000
26.21429
Upper
23.6049
28.8237
6. a) The null hypothesis for this situation would be that there was no difference in the number of questions answered correctly
between the control condition and stereotype threat condition (H0: µc= µs). The alternative hypothesis is that there is a difference in the
number of questions answered correctly between the control condition and stereotype threat condition (H1: µc≠ µs).
b) An independent t-test needs to be conducted in this situation. Within this study there were 23 participants who were randomly
assigned to either the control or experimental condition. In order to compare the means of the two groups an independent samples ttest must be performed.
c) An independent samples t-test was performed and a significant difference was found, t (21) = 2.334, p < 0.032 (two tailed),
therefore reject the null hypothesis and conclude that the participants in the control condition (M = 9.64, SD= 3.17) answered
significantly more of the questions correctly compared to the stereotype condition (M=6.83, SD= 2.52).
Evidence from SPSS:
Group Statistics
condition2
N
Mean
Std. Deviation
Std. Error Mean
control
11
9.6364
3.17089
.95606
stereotype threat
12
6.8333
2.51661
.72648
CorAnswr
\
LAB ASSIGNMENT # 2
7
Independent Samples Test
Levene's Test for
t-test for Equality of Means
Equality of Variances
F
Sig.
t
df
Sig. (2-
Mean
Std. Error
95% Confidence Interval
tailed)
Difference
Difference
of the Difference
Lower
C
Equal variances
oassumed
1.644
.214
Upper
2.359
21
.028
2.80303
1.18840
.33162
5.27444
2.334
19.095
.031
2.80303
1.20076
.29065
5.31541
r
A
nEqual variances not
sassumed
w
r
Download