Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 3

Some Data Preparation Techniques

A. Compute Variable
Transform → Compute Variable → Input the brief name of the target variable
inside “Target Variable” (space is not allowed) box → Click “Type & Label” → Input
the Name of the target variable inside the “Label” box or you may opt to copy the
brief name as name of the variable by clicking the “Use expression as label” →
Select the appropriate Mathematical equation needed to compute for the variable →
Plug the needed variables to the selected Mathematical equation → Click Ok

B. Select Cases
Data → Select Cases → check If condition is satisfied → click “If” → Plug the
variable/s where selection will be effected (e.g. Gender) to the variables box and
select the appropriate selection condition (e.g. Gender = 1 if you want to select
“Males”) → Continue → Under “Output”, select which actions you need to take (i.e.
filter out selected cases, copy selected cases to a new dataset, or delete unselected
cases) → Ok

C. Recode Variables

1. Recode Variable into Same Variable


Transform → Recode into same variable → Plug the variable of interest to the
variables box → Click old and new values → specify the old and the new values →
Continue → OK

2. Recode Variable into a Different Variable


Transform → Recode into different variables → Specify the Name and Label of
the Output variable → Click Change → Click Old and New Values → Specify
necessary changes in the old and new values → Continue → OK

How to Report Quantitative Evidence

Bivariate Correlation
To investigate if there is a statistically significant association between V1 and V2, a
correlation was computed. The Pearson’s r was calculated, r(n) = value of r, p < .001.
The direction of the correlation is positive, which means that V 1 tend to have a higher V 2
and vice versa.
Independent Samples t-Test
Table 1
Comparison between students who underwent the intervention (Section Rizal) and
students who did not undergo the intervention (Section Mabini) in terms of their
summative test results (Rizal = 35; Mabini=35)
Variable M SD t df p
Summativ 2.974 68 .004
e Test
Rizal 90.40 5.558
Mabini 86.31 5.930
Table shows that intervention group (section Rizal) scored significantly higher than the
traditional group (section Mabini) in the summative test (p=.004). This means that the
summative test scores of section Rizal (90.40) is significantly higher than section Mabini
(86.31). The effect size is .71 which suggests that the intervention group is at 71%
advantage over the traditional group.

Notes on Test of Equality of Variances:


1. The Levene’s test examines whether the variances of the two groups are equal.
2. The p value under Levene’s test tells whether there is a significant difference in the
variances of the two groups.
3. Case 1: p value higher than 0.05: there is NO SIGNIFICANT difference between the
two variances. Therefore, the variances are EQUAL (groups comparable).
Therefore, the assumption of equality of variance is satisfied. Therefore, use the
“Equal variances assumed” value
Case 2: A P value is equal to or lower than 0.05: there IS A SIGNIFICANT
difference between the two variances. Therefore, the variances are NOT EQUAL.
Therefore, the assumption of equality of variance is NOT SATISFIED. Therefore,
use the “Equal variances not assumed” value
4. The p-value of the t-test tells us whether there is a significant difference between the
two groups we are comparing in terms of a given score/measure.

Analysis of Variance (ANOVA)


A statistically significant difference was found among three levels of IV in DV, F(degrees
of freedom between, degrees of freedom within) = F value, p=p value. Table Z shows
that the mean sores of the DV is ________ for samples of IV level1, __________ for
samples of IV level 2, and ___________ for samples of IV level 3. Post hoc Tukey HSD
Test indicate that (discuss levels of IV that return statistically significant different means)
Table 2
A One-Way Analysis of Variance comparing the performance of sections Rizal, Mabini,
and Aguinaldo in the Summative Test.

Summative Test
Section n
M SD
Rizal 35 90.40 5.558
Mabini 35 86.31 5.930
Aguinaldo 35 86.94 5.800
Total 105

A statistically significant difference was found among sections Rizal, Mabini, and
Aguinaldo in terms of their scores in the summative test, F(2, 102) = 5.097, p < .01.
Table 2 shows that the mean score in the summative test is 90.40 for section Rizal,
86.31 for section Mabini, and 86.94 for section Aguinaldo.

Bivariate Linear Regression


Bivariate Linear Regression was conducted to investigate whether the students’ scores
in diagnostic test significantly predict their scores in the summative test. Results
revealed that this predictor accounts about 20% of the variance of the predicted
variable. Further, ANOVA results showed that this prediction is significant, F=25.716,
p<.05.

Table 3
Variable B SE β
Diagnostic Test .529 .104 .447
42.654 8.935
Note. R2=.20; F=25.716, p<0.5

The beta coefficient presented in Table 3 shows that a unit increment in the diagnostic
test translates to 0.447 units increase in the summative test.

Reference: Morgan, G.A., Leech, N.L., Gloeckner, G.W., Barrett, K.C. (2004). SPSS for
Introductory Statistics: Use and Interpretation (2nd Edition). Lawrence Erlbaum
Associates, Publishers

You might also like