You are on page 1of 13

Lab practical – PSY3000 Summer 2016

Materials allowed
1) Lab handouts
2) Phones, Tablets, and Computers
3) Calculator
4) Internet
5) Book
6) Class notes
7) Anything you can bring to bear 

Instructions

There are five research questions below. For each question, identify what variables you are
being asked to analyze and conduct the appropriate statistical test. You will need to consider the
level of measurement (nominal, scale, ordinal) of the variables, the number of groups or levels,
and whether the variable is within- or between-subjects.

Turn in your output showing the analyses and graphs, and write your interpretation/summary of
what the results of each analysis mean in words, on this handout. You can earn an additional 5
percentage points of extra credit if you write your test results in APA format. You can earn an
additional 5 percentage points of extra credit if you write your test results in APA format.
If you choose not to write your results in APA format, include at minimum relevant test
statistics (like t-values, F-values, and p-values) so that I can see how you reached your
conclusions.

Finally, if the results of your analysis require a post hoc test to draw a conclusion, include this in
your output and write-up).

Please answer the following questions using the dataset provided. You will not need to enter data
or change variable characteristics.

Here is the story behind the dataset: A researcher measured overall emotional reactivity and
anxiety in a group of 68 participants. In addition, participants were assigned to one of three
conditions: neutral, surprising, or angry video. Approximately half of the participants were only
given audio of the video, and the other half were shown the video with both audio and images.
Participants’ positive and negative affect were measured before (Time 1) and after the video
(Time 2).
Questions

1) Did positive affect (pos_aff) change from time 1 to time 2? How about negative affect
(neg_aff) from time 1 to time 2? Report the 95% CI. Also report the effect size if
appropriate. If writing your results in APA format, include the relevant descriptive
statistics.

We wanted to understand if there was a significant difference between positive affliction in time
1, (M= 29.30, SD = .999) and positive affliction in time 2 (M=23.80, SD = 1.101) using a
dependent 2 sample t-test and found the results significant; (t(67)= 7.625, p > .001). Another
dependent 2 sample t-test was used to test the difference between the negative affliction in time 1
(M=15.49, SD = .880) and the negative affliction in time 2 (M= 14.55, SD = .754) and there was
not enough evidence to support a significant difference; (t(67)=1.509, p = .136). I’m 95%
confident that the true difference between the means in time 1 and time 2 regarding positive
affliction is between 4.059 and 6.938.
Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

Pair 1 pos_aff_Time1 29.299 68 8.2336 .9985

pos_aff_Time2 23.800 68 9.0771 1.1008


Pair 2 neg_aff_Time1 15.493 68 7.2573 .8801

neg_aff_Time2 14.547 68 6.2155 .7537


Paired Samples Test

Paired Differences

95% Confidence Interval of

Std. Error the Difference

Mean Std. Deviation Mean Lower Upper t df Sig. (2-tailed)

Pair 1 pos_aff_Time1 -
5.4985 5.9464 .7211 4.0592 6.9379 7.625 67 .000
pos_aff_Time2
Pair 2 neg_aff_Time1 -
.9456 5.1661 .6265 -.3049 2.1961 1.509 67 .136
neg_aff_Time2

2) Did the participants’ negative affect at time 2 differ by sex?

We wanted to know if there was a difference between males (M=16.31, SD=1.383) and females
(M= 13.07, SD=.690) in regards to their scores in the negative affliction in time 2 using an
independent 2 sample t-test and the results were significant; (t(66)= 2.201, p= .031). I’m 95%
confident that the true variation of the mean difference falls between .301 and 6.178.

Group Statistics

sex N Mean Std. Deviation Std. Error Mean

neg_aff_Time2 Male 31 16.310 7.6985 1.3827

Female 37 13.070 4.1945 .6896

Independent Samples Test

t-test for Equality of Means

95% Confidence Interval of


Mean Std. Error the Difference

t df Sig. (2-tailed) Difference Difference Lower Upper

neg_aff_Time2 Equal
variances 2.201 66 .031 3.2394 1.4717 .3010 6.1778
assumed
3) Examine how negative affect at time 2 differed by sex and picture groups. Is there an
interaction? If any of your results are significant, include the relevant descriptive
statistics in your write-up.

In order to study the relationships between sex (male vs. female) and picture groups (audio only
vs. audio plus images) and the interaction effect regarding scores in the negative affliction in
time 2 we used a 2 x 2 between-subjects factorial ANOVA. There was a main effect of sex,
(F(1,64)= 4.759, p = .033), such as males tended to score higher than females regardless of the
picture group and when calculating a pairwise comparison on sex they were consistent with our
significant main effect findings. However, there was no main effect regarding whether that
participants were shown only audio or audio paired with images, (F(1,64)=.053, p=.819) nor
was there any significant interaction between sex and picture group regarding the scores in
negative affliction in time 2, (F(1,64)=.022, p=.883) so no pairwise comparisons were computed
for these results. Therefore, the main effect by being male resulted in higher scores in the second
timed study of negative afflictions regardless of what picture group they were in, which 7% of
variability could be explained by this main effect, (R² = .070).
Tests of Between-Subjects Effects
Dependent Variable: neg_aff_Time2

Type III Sum of


Source Squares df Mean Square F Sig.

Corrected Model 180.059a 3 60.020 1.595 .199


Intercept 14487.186 1 14487.186 384.992 .000
sex 179.096 1 179.096 4.759 .033
picture 1.993 1 1.993 .053 .819
sex * picture .821 1 .821 .022 .883
Error 2408.310 64 37.630
Total 16978.320 68
Corrected Total 2588.369 67

a. R Squared = .070 (Adjusted R Squared = .026)


Pairwise Comparisons
Dependent Variable: neg_aff_Time2

95% Confidence Interval for

Mean Differenceb

(I) sex (J) sex Difference (I-J) Std. Error Sig.b Lower Bound Upper Bound

Male Female 3.264* 1.496 .033 .275 6.253


Female Male -3.264* 1.496 .033 -6.253 -.275

Based on estimated marginal means


*. The mean difference is significant at the .05 level.
b. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustments).

3. sex * picture
Dependent Variable: neg_aff_Time2

95% Confidence Interval

sex picture Mean Std. Error Lower Bound Upper Bound

Male Audio Only 16.250 1.534 13.186 19.314

Audio plus images 16.373 1.584 13.209 19.537


Female Audio Only 12.765 1.488 9.793 15.737

Audio plus images 13.330 1.372 10.590 16.070

4) Did average anxiety differ by condition? If so, which condition groups were different
from each other, and which were not different from each other? If writing your results in
APA format, be sure to include the relevant descriptive statistics.

We were interested seeing if there was a difference in anxiety when exposed to the various
conditions (neutral, surprising, and angry videos) we used a 3 x 1 one-way factorial
ANOVA and found significant results; (F(2,65)=4.115, p=.021). When computing a
Turkey’s Post-Hoc test we found significant results between neutral (M= 37.86, SD = 2.244)
and angry (M= 45.74, SD= 1.843) videos (p= .030). I’m 95% confident that the true effect
size between neutral and angry videos falls between .65 and 15.10. There was no significant
results when comparing surprising (M =34.13 , SD = 2.263) videos to either neutral or angry
videos.
ANOVA
anxiety

Sum of Squares df Mean Square F Sig.

Between Groups 840.199 2 420.100 4.115 .021


Within Groups 6636.330 65 102.097
Total 7476.529 67

Multiple Comparisons
Dependent Variable: anxiety
Tukey HSD

Mean Difference 95% Confidence Interval

(I) condition (J) condition (I-J) Std. Error Sig. Lower Bound Upper Bound

Neutral Videos Surprising Videos -.962 3.013 .945 -8.19 6.27

Angry Videos -7.875* 3.013 .030 -15.10 -.65


Surprising Videos Neutral Videos .962 3.013 .945 -6.27 8.19
Angry Videos -6.913 2.980 .060 -14.06 .23
Angry Videos Neutral Videos 7.875* 3.013 .030 .65 15.10

Surprising Videos 6.913 2.980 .060 -.23 14.06

*. The mean difference is significant at the 0.05 level.

Descriptives

condition Statistic Std. Error

anxiety Neutral Videos Mean 37.86 2.244

95% Confidence Interval for Lower Bound 33.20


Mean Upper Bound 42.53

5% Trimmed Mean 37.00

Median 34.50

Variance 110.790

Std. Deviation 10.526

Minimum 28

Maximum 64

Range 36
Interquartile Range 15

Skewness 1.187 .491


Kurtosis .396 .953

Surprising Videos Mean 38.83 2.263

95% Confidence Interval for Lower Bound 34.13


Mean Upper Bound 43.52

5% Trimmed Mean 38.31

Median 37.00

Variance 117.787

Std. Deviation 10.853

Minimum 24

Maximum 63

Range 39

Interquartile Range 12

Skewness .850 .481

Kurtosis .589 .935

Angry Videos Mean 45.74 1.843

95% Confidence Interval for Lower Bound 41.92


Mean Upper Bound 49.56

5% Trimmed Mean 45.52

Median 44.00

Variance 78.111

Std. Deviation 8.838

Minimum 32

Maximum 63

Range 31

Interquartile Range 11

Skewness .661 .481

Kurtosis -.271 .935


5) Are emotional reactivity and negative affect at time 1 associated? If so, perform a
regression analysis predicting negative affect at time 1 using emotional reactivity as the
predictor. State whether or not the results of your regression equation are significant.
Include the regression equation in your answer (interpreting it in plain language). What
was your R-squared value? Provide an interpretation.
Finally, what is the predicted negative affect for someone who has an emotional
reactivity score of 47.5?

We were interested in seeing if emotional reactivity (M = 37.53, SD = 9.394) was a


good predictor for negative affliction (M=15.49, SD = 7.257) we found that 13.2 % of
the variability in negative affliction could be explained by emotional reactivity (R² =.132,
SD = 6.812), which was consistent with our significant correlation strength results; (R =
.363, p= .001). Using the prediction equation, y’ =4.96+0.28(x), we predict that a value
of 47.5 in emotional reactivity would produce a score of 18.26 in negative affliction in
time 1.
4.96+0.28(47.5) = 18.26

Descriptive Statistics

Mean Std. Deviation N

neg_aff_Time1 15.493 7.2573 68


emotional_reactivity 37.53 9.394 68
Correlations

emotional_reacti
neg_aff_Time1 vity

Pearson Correlation neg_aff_Time1 1.000 .363

emotional_reactivity .363 1.000


Sig. (1-tailed) neg_aff_Time1 . .001
emotional_reactivity .001 .
N neg_aff_Time1 68 68

emotional_reactivity 68 68

Model Summary

Adjusted R Std. Error of the


Model R R Square Square Estimate

1 .363a .132 .119 6.8123

a. Predictors: (Constant), emotional_reactivity

ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression 465.897 1 465.897 10.039 .002b

Residual 3062.850 66 46.407

Total 3528.746 67

a. Dependent Variable: neg_aff_Time1


b. Predictors: (Constant), emotional_reactivity

Coefficientsa

Standardized
Unstandardized Coefficients Coefficients 95.0% Confidence Interval for B

Model B Std. Error Beta t Sig. Lower Bound Upper Bound

1 (Constant) 4.958 3.426 1.447 .153 -1.882 11.798

emotional_reactivity .281 .089 .363 3.169 .002 .104 .458

a. Dependent Variable: neg_aff_Time1

You might also like