USING SPSS

ACTIVITY 1. RELATED OR DEPENDENT SAMPLES

EXAMPLE 1. A study on the high crime rate in Manila was conducted. The number of crimes in each of eleven
(11) sample areas during the six (6) months was recorded. Then a Neighborhood Watch Program (NWP) was
launched. The number of crimes before and after the implementation of the program was recorded. Test at
0.05 level of significance the hypothesis that the program has decreased the number of crimes committed.
(IS THERE A SIGNIFICANT DIFFERENCE IN THE NUMBER OF CRIMES BEFORE AND AFTER THE IMPLEMENTATION
OF THE NEIGHBORHOOD WATCH PROGRAM?)
AREA NUMBER OF CRIMES
BEFORE AFTER
1 14 9
2 9 7
3 5 3
4 5 6
5 19 8
6 12 13
7 8 3
8 9 4
9 10 5
10 8 9
11 11 7

Step 1. Ho: __________________________________________________________________________

Ha: _________________________________________________________________________
Step 2.Use SPSS
 Open variable view
o Name – no space, no special characteristics
o Type – numeric(N), string (L) if letters
o Label –can be omitted or same as in Name
Name Type width Decimals Label Values Missing Column Align Measure Role
s
1
2
 Type Under name for 1 – BEFORE, 2-AFTER
 click data view and fill in the data of the number of crimes (as in the table above)
(TEST FOR NORMALITY – SKEWNESS AND KURTOSIS TEST)
 click the following:analyze, descriptive statistics, descriptives
 highlight “before” then click the arrow to transfer to variable box; do the same for “after”
 click options, click skewness and kurtosis then continue, then ok
OUTPUT: Divide statistic with standard error. If the result iswithin -1.96 to 1.96 (at alpha 0.05)
then the data is normally distributed.
(ANOTHER TEST FOR NORMALITY – KOLMOGOROV-SMIRNOV TEST)
 click analyze, nonparametric test, legacy dialog, one-sample k-s..
 transfer “before” and “after” to test variable test (highlight, click arrow)
 OUTPUT: If the p-value (assymp sig) is greater than the alpha (0.05) then “The test distribution is
normal”
(PAIRED T-TEST)
 Click: analyze, compare means-paired sample t-test
 Transfer paired variables through the arrow, then ok
 OUTPUT (refer to last table)
o what is the t-value? (that is the t-computed)
o what is in sig (2-tailed)? That is the p-value.
Step 3. Compare it with the alpha, then make a decision: If the p-value ≤ the alpha which is 0.05, then REJECT
Ho. Otherwise, do NOT REJECT Ho.
Step 4. Make your conclusion (where the Ho is the claim) : With 0.05 level of significance, there
is________________________________________________________________________________________
ACTIVITY 2. UNRELATED OR INDEPENDENT SAMPLES

EXAMPLE 1. Mean corpuscular hemoglobin in microorganisms calculated as Hemoglobin content (g/1000 cc)
divided by Red Blood cells (millions per cu mm)was determined at the age of 10 months for two groups of
babies. The groups were comparable for all known factors that might lead to iron deficiency. In one group,
the umbilical cords WERE CLAMPED IMMEDIATELY AFTER BIRTH, and in other, cords WERE CLAMPED AFTER
THE PLACENTA BEGAN TO DESCEND INTO THE VAGINA. Is there a significant difference in the mean
corpuscular hemoglobin of babies when they are grouped as to clamping of umbilical cords? Test this at
0.05 level of significance given the following sets of data in two groups:

Mean corpuscular Hemoglobin in Microorganisms at 8 to 10 months of age

Grou Clamped 17.1 25.6 31.1 17.3 17.3 18.1 20.9 22.2 15.8 29.8 23.9 27.8 17.9 20.0
pA
Grou Not 13.5 20.3 26.5 23.3 27.1 30.5 26.3 36.1 30.0 25.0 28.7 21.1
pB clamped

Step 1. Ho: __________________________________________________________________________

Ha: _________________________________________________________________________
Step 2. Use SPSS
 Open variable view
o Name – type GROUP (1)
o Type – numeric(N), string (L) if letters
o Label –can be omitted or same as in Name
o Values – Click values
 Value : 1, label: clamped, the add
 Value : 2, label: not clamped, the add, then ok
o Name – type MCH (2)
Name Type width Decimals Label Values Missing Column Align Measure Role
s
1 Group
2 MCH
 Click data view
 Type under group, 1 (1-14) and type against each number in column 2 the numerical data under group
A-clamped
 Type 2 (15-26) and type against each number in column 2 the numerical data under group B-not
clamped
 Check accuracy of data (A 1)
(TEST FOR NORMALITY – SKEWNESS AND KURTOSIS TEST)
(ANOTHER TEST FOR NORMALITY – KOLMOGOROV-SMIRNOV TEST)

(INDEPENDENT T-TEST)
 Click: analyze, compare means-independent sample t-test
 Transfer test variables
o Group to group variables then define groups: Group 1 is 1 and group 2 is 2
o MCH to test variables then ok
 OUTPUT (refer to last table)
o Choose equal variances assumed, when the Levene’s Test of Equality is significant;
Otherwise choose equal variances not assumed
 o What is the t-value?
o what is in sig (2-tailed)? That is the p-value.
Step 3. Compare it with the alpha, then make a decision: ____________________________
Usually in the thesis, the table is presented this way:
Variables t-value p-value Alpha Decision Conclusion
Clamped versus not clamped umbilical -1.844 .081 0.05 FR H
0
Not significant
cords
Step 4. Make your conclusion (where the Ho is the claim) : With 0.05 level of significance, there
is________________________________________________________________________________________
Exercises (use SPSS)

1. The plant manager of a company that manufactures office equipment believes that worker productivity is a
function of, among other things, the design of the job, which refers to the sequence of workers' movements.
Two designs are being considered for the production of a new type of ergonomics (human engineering to
design and arrange things so that people will interact most effectively and safely) computer desk. To help
decide which should be used, an experiment was performed. Twenty (20) randomly selected workers
assembled the desk using design A and 16 workers assembled the product using design B. The assembly time
in minutes was recorded and exhibited below. The plant manager would like to know whether the assembly
time of the two designs differ. Test this at 5% significance level.

Data on Assembly Times in Minutes

Design A Design B
6.8 6.5 6.1 6.4 7.1 5.2 6.6 7.1 7.1
5.2 7.1 7.7 7.6 7.0 6.7 5.9 4.2 8.5
6.0 6.6 5.2 6.2 5.9 7.0 5.3 6.6 4.5
6.1 7.9 6.1 6.3 4.6 4.9 5.7 4.2 7.0

--------------------------------------------------------------------------------------------------------------------------------------------------

2. Fifteen individuals were given a test on self-esteem before and after participating in a course intended to
enhance self-esteem. The result are as follows:
Individual Change in Self-Esteem
Before After
1 90 93
2 91 92
3 93 95
4 89 88
5 85 88
6 87 89
7 83 84
8 80 83
9 84 85
10 82 83
11 75 77
12 81 86
13 72 74
14 70 70
15 71 76
Test the hypothesis at 0.05 whether the course has increased the individual’s self-esteem.
1. Ho:______________________________________________________________
Ha: ______________________________________________________________
2. Test statistic:
3. Decision:
p-value: ______
(attach SPSS outputs)
4. Conclusion