Psych 248 – Lab SPSS Assignment #2 – Hypothesis Testing Prof.

Hamid summer 1, 2019

DUE: Monday July 1st, 2019 at 11:59 PM

Answer all questions on a separate word document, save and send via online blackboard
submission.doc, .docx or .pdf. MAKE SURE YOU INCLUDE EVERYTHING THAT IS ASKED FOR!
Everyone should hand in their OWN assignment, even if you work with other people – your responses
MUST be your own! Anyone who hands in assignments where responses are word-for-word exactly the
same will receive a ZERO on the assignment.

About the dataset:

Elmo: brain-building or brain-draining?

The television series Sesame Street is concerned mainly with teaching preschool skills to children age 3-
5, with special emphasis on reaching economically disadvantaged children. The show is designed to hold
young childrens' attention through action oriented, short duration presentations teaching specific
preschool cognitive skills and some social skills. Each show is one hour and involves much repetition of
concepts within and across shows.

Does Sesame Street help economically disadvantaged children 'catch-up' with economically advantaged
children? In the early 1970s, researchers at Educational Testing Service (the company that runs the SAT)
ran a study to evaluate Sesame Street. The researchers sampled children representative of economically
advantaged and disadvantaged populations from five different sites in the United States.The children
were tested on a variety of cognitive variables, including knowledge of body parts, knowledge about
letters, knowledge about numbers, etc., both before and after viewing the series.*

A sample of the data used to evaluate the impact of Sesame Street is available in the file
“Lab2.Sesamest.sav” on Blackboard (Folder: Assignments Assignment 2) posted along with this
assignment. The names of variables are shown in the code book on the last page of this assignment.

(*Note: Description courtesy of Jerry Reiter, stats professor at Duke University & modified from Jennifer
Hill’s Causal Inference class at NYU)
 1) Open the dataset, “Lab2.Sesame.sav”

How many variables are in the dataset? ___9___

How many cases are in the dataset? ___2___

For following questions you will be asked to conduct a hypothesis test analysis in SPSS. You
must decide whether the situation fits a one sample t-test, independent samples t-test or a
paired-sample t-test.

2) A researcher is not sure if the sample is representative of the entire population of pre-
school students in the US in terms of their PPVT scores (peabody variable). Run a test to see
if the sample in the dataset is representative of the population of pre-school students in the
US (µ = 49, please note this is not really the population mean, it was made up for the
purpose of this assignment only!).

Copy and paste the output of the test results into your assignment write-up document, state the
descriptives of the sample (mean, standard deviation, standard error) as well as the critical t-
value, and answer/complete the following:

One-Sample Statistics

N Mean Std. Deviation Std. Error Mean

peabody 240 46.47 16.039 1.035

One-Sample Test

Test Value = 0

95% Confidence Interval of the

Difference

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

peabody 44.883 239 .000 46.467 44.43 48.51

  What kind of t-test did you choose to run and why?
I chose to use a one sample-test because two different means are being compared
(population and sample means) to see if there is a significant difference between
them.
 One of the assumptions of t-tests is that the outcome variable is approximately
normal, test for this in Analyze  Descriptive Statistics  Explore and choose the
option “plot normality test”.
Tests of Normality

Kolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

peabody .131 240 .000 .913 240 .000

a. Lilliefors Significance Correction

 o Interpret the results of the Shapiro-Wilk test and determine if it would have
been better to conduct a non-parametric test or not.
o What is the correct non-parametric test alternative to the t-test you conducted?
o Null means they are
 State the null and alternative hypothesis for the t-test you ran.
 Write an APA style conclusion (~2 sentences) whether the test was significant, and
include the t-statistic that was calculated, df, p-level, and 95% confidence interval.

3) Perform two hypothesis tests, one for the regular=1 (treatment) group and one for the
regular = 0 (control) group, for the pre-test/post-test variables prenumb and postnumb. To do
the analysis by group, go to Data  Split File  click on “organize output by groups” and
place the grouping variable in the box.

Copy and paste your output into your word document, state the descriptives of the two
samples for both variables (mean, standard deviation, standard error), and then answer the
following:

Paired Samples Testa

Paired Differences

95% Confidence Interval of the

Difference

Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-ta

P
prenumb - postnumb

i
-10.478 10.073 .739 -11.936 -9.021 -14.187 185
r

a. regular = treatment
 Paired Samples Statisticsa

Mean N Std. Deviation Std. Error Mean

Pair 1 prenumb 21.96 186 10.787 .791

postnumb 32.44 186 12.202 .895

a. regular = treatment

Paired Samples Correlationsa

N Correlation Sig.

Pair 1 prenumb & postnumb 186 .622 .000

a. regular = treatment

 What kind of t-test(s) was(/were) used for this analysis? Why?

Paired sample test
 Describe the descriptives for the samples (means, standard deviations, standard error
and the mean differences)
 State the null and alternative hypotheses.
 Write an APA style conclusion with 2-4 sentences. Include the t-statistics, and
whether the mean differences are significant or not (state p-level and df.) and the 95%
CI’s for each t-test that you ran.

4) Conduct a hypothesis test for the regular =1 vs. regular = 0 (consider it “treatment” vs.
“control” group) for the post-test variable postlet.
Copy and paste the output from this test into your document, and answer the following:

Paired Samples Statisticsa

Mean N Std. Deviation Std. Error Mean

Pair 1 prelet 16.38 186 8.948 .656

postlet 29.60 186 12.997 .953

a. regular = treatment

Paired Samples Correlationsa

N Correlation Sig.

Pair 1 prelet & postlet 186 .546 .000

 a. regular = treatment

Paired Samples Testa

Paired Differences

95% Confidence Interval of the

Difference

Mean Std. Deviation Std. Error Mean Lower Upper t df Sig.

Pair prelet - postlet

-13.220 11.046 .810 -14.818 -11.623 -16.323 185
1

a. regular = treatment

 How many participants are in each sample group?

 What kind of t-test did you choose to run? Why?
 State the means of each group for postlet and their respective standard deviations and
standard errors as well as the mean difference.
 State the null and alternative hypothesis and state the critical t* value (∝ = .05; two-
tail test).
 According to Levene’s test for Equality of Variances – can you assume the treatment
vs. control group have equal variances for the postlet scores? Explain why or why
not. Discuss which row you should look at for the correct test statistics in the output.
 Write an APA style conclusion including the t-statistic, df, and 95% confidence
interval.

Make sure you have answered all questions on a separate document, and have included any
output asked for and the graphs. HAND IN ONE DOCUMENT ONLY – no dataset, no output
file, no syntax. Everyone is responsible for handing in their own work and using their own
words, even if you worked with others on the assignment.
 Codebook with variable names for Sesame Street data

id : subject identification number

gender: male=1, female=0

regular: frequency of viewing: 0=rarely watched the show (control), 1= watched once/week or
greater (treatment)

prebody : pretest on knowledge of body parts (range of scores: 0-32)

prelet : pretest on letters (range of scores: 0-58)

prenumb : pretest on numbers (range of scores: 0-54)

postbody : posttest on knowledge of body parts (range of scores: 0-32)

postlet : posttest on letters (range of scores: 0-58)

postnumb : posttest on numbers (range of scores: 0-54)

peabody: mental age score obtained from administration of the Peabody Picture Vocabulary test
(PPVT) as a pretest measure of vocabulary maturity (range of scores: 8-99)