MODULE 5 MEASURES OF COMPARISON

LESSON 1 ANALYSIS OF COMPARISON

Learning Objectives:

At the end of the lesson, you should be able to:

1. distinguish between the uses of the different statistical measures for comparison:
t-test and Analysis of Variance;
2. understand the application and uses of the t-test; and
3. understand the application and uses of analysis of variance.

Keywords and Phrases:

 t-test
 analysis of variance (ANOVA)
 independent samples
 dependent samples

Introduction

So far in hypothesis testing, a researcher not only deals with tests of relationships

but also with differences. When different groups are used in research, these groups usually

represent a research variable. When a researcher tests for group differences on a particular

measure, he is also assessing comparison or relationship.

T-test
Significance of the Difference Between Two Means for Independent Samples.
Consider drawing a very large number of pairs of samples, each from one of two
“treatment population” and study. Then we compute the difference in sample means x́ 1- x́ 2. This

94
is a theoretical sampling distribution and its mean is equal to the difference in population means,
x́ 1−x́ 2.

For two independent samples, the standard deviation of this sampling distribution
is called the standard error of the difference of the two means, written as,
δ x́ 1− x́ 2

Let’s assume that we are going to conduct an experiment using an experimental

and a control group. We will use x́ 1as the mean of the control group and x́ 2, the mean of the
experimental group with N 1 and N 2cases, respectively, drawn from a normally distributed
population with means μ1 and μ2 and variances δ 21 and δ 22.

The assumption is that the samples are drawn from populations with equal
variances δ = δ 21=δ 2.
2
1

This is known as homogeneity of variance assumption (homoscedasticity). If this

assumption is true, data from the two samples can be combined to obtain an estimate of the
population variance ( s2) as:

N1
2
S = ∑ ¿¿ ¿
i=1

Or when s1 and s2 are given:

S 1 (N −1)+ S 2 ( N −2)
2 2

s2 =
N 1+ N 2 −N

The standard error of the difference between the two means is given by:

s 2+ s 2
s x́ − x́ =
1 2
√ N1 N2

95
To obtain the t-ratio, the difference between the mean is divided by the estimate of the standard
error,

x́1− x́2
t=
s x́ −x́
1 2

Another formula for calculating s2 (pooled variance) when raw scores are given is:

s2 =¿ ¿
Sample Calculation:
At the start of a semester, a teacher draws a sample of students in his mathematics class
and randomly assigned 15 of them to an experimental group and 15 to a control group. The
teacher taught the experimental group (Group 1) with Computer-Aided Instruction (CAI) and the
control group with the traditional method of teaching. At the end of semester, a mathematics
achievement test was given to both groups. The results are given below. On the basis of the data,
should the teacher conclude that the experimental method is more effective than the traditional
method?
Experimental group Control group
35 25
45 32
36 40
20 28
25 22
35 25
40 20
45 50
27 20
32 18
20 12
18 40
15 15
12 10
10 8

Hypothesis:
H 0 : μ1 ¿ μ 2 or μ1−μ 2=0

96
The assumption is made that x́ 1 ,´x 2 are means of two samples drawn from a normally distributed
population means μ1and μ2 and variances δ 21and δ 22 .

μ1 ≠ μ2 or μ1- μ2 = 0

Level of significance: α =0.05

Region of rejection: -2.048
And t > 2.048, df = 28

Computation:

X1 X2 X 21 X 22
35 25 1225 625
45 32 2025 1024
36 40 1296 1600
20 28 400 784
25 22 625 484
35 25 1225 625
40 20 1600 400
45 50 2025 2500
27 20 729 400
32 18 1024 324
20 12 400 144
18 40 324 1600
15 15 225 225
12 10 144 100
10 8 100 64
∑ X 1 = 415 ∑ X 2 = 365 ∑ X 3 = 13367 ∑ X 4 = 10899

x́ 1 = 27.7
x́ 2 = 24.3
∑ x 21 ∑ x12
N 1 = 15 N 2= 15

The unbiased estimate of the variance is:

97
 s2 =¿ ¿

(415)2 (365)2
s2 = [ 13,367−
15 ][
+ 10,899−
15 ]
15+15−2

[ 13,367−11481.67 ] + [ 10,889−8,881.67 ]
s2 =
28

s2 = 139.38

The t-ratio is:

X́ 1− X́ 2
t= s2 + s2
√ N1 N 2

27.7−24.3
t= 139.38 139. .38
√ 15
+
15

3.4
t=
√ 27.876
3.4
t= 5.28 = 0.64

Decision Rule:

The tabular value of t for df 28 at 0.05 level is 2.048. The observed value of 0.46
indicates that the difference between the means was not large enough for us to reject
98
the null hypotheses at the 5% level. We can conclude that the two methods of teaching
do not exert a differential effect on the achievement of students.

SIGNIFICANCE OF THE DIFFERENCE BETWEEN TWO MEANS FOR

INDEPENDENT SAMPLES IN SPSS

To run independent sample t-test in SPSS, select Analyze, Compare Means,

Independent-Samples T-test… from the data editor window.

The Independent-Samples T-test window will appear after clicking on the

independent-samples t-test option. Select the test variable (scores) and put them on the

right hand side of the test variable list. Select the “grouping variable: to define groups,
then click on the “ok” button. (In this example, the two groups are the students who
used experimental method and students who utilized traditional method).

99
The following table will show the result.

Group Statistics

Group N Mean Std. Deviation Std. Error Mean

Scores Experimental Method 15 27.6667 11.60460 2.99629

Traditional Method 15 24.3333 12.00397 3.09941

100
 Independent Samples Test

Levene's Test for

Equality of
Variances t-test for Equality of Means

95% Confidence
Interval of the

Sig. (2- Mean Std. Error Difference

F Sig. t df tailed) Difference Difference Lower Upper

Scores Equal .072 .791 .773 28 .446 3.33333 4.31093 -5.49721 12.16388
variances
assumed

Equal .773 27.96 .446 3.33333 4.31093 -5.49766 12.16433

variances 8
not
assumed

MORE SPSS APPLICATION – t-TEST FOR INDEPENDENT SAMPLES

Example

A researcher is interested whether there is a significant difference in the teaching

performance of male and female teachers. The researcher randomly selected 24
teachers, noted their gender, and evaluated their teaching performance. Gender is
coded 1 for male and 2 for female. The data are shown in Table 2.

Table 2. The Gender and Teaching Performance of 24 Teachers

Case Gender Teaching Performance

1 1 98
2 1 95

101
 3 1 92
4 1 94
5 1 89
6 2 79
7 2 82
8 2 88
9 2 85
10 1 94
11 2 83
12 1 97
13 2 80
14 1 91
15 2 82
16 1 90
17 2 81
18 2 83
19 2 84
20 2 85
21 2 85
22 2 80
23 2 90
24 1 97

Step 1. Enter your data in the SPSS Data Editor.

102
Figure 2.1 SPSS Data Editor after independent samples t-test data entered in SPSS

Data Editor

Step 2. Run the Independent Samples t-test.

 Click on Analyze.
 Click on Compare Means.
 Click on Independent-Samples T-Test
 A dialog box will open with a list of your variables in the white rectangle on the
left hand side. Click on the dependent variable, and then click on the small black
arrow to the left of the white box headed Test Variable(s). This will place the
dependent variable (in this case Teaching Performance) in the box.
 Click on the independent variable and then click on the small black arrow to the
left of the white box headed Grouping Variable. This will place the independent
variable (in this case Gender) in the box (see Figure 2.2)

103
Figure 2.2 Entering the test and grouping variables
 Click on the Define Groups button.
 The dialog box shown in Figure 2.3 will open. Enter the two values of the

independent variable in the white boxes. Because in our example males are

coded 1 and females 2 (see Table 2), these are the two values shown in Figure

2.3.

104
Figure 2.3 Entering the levels of the independent variable

 Click on Continue.
 Click on OK.

The output will be as follows

Group Statistics
Gender Std.
  Std. Error
N Mean Deviation Mean
TeachingPerformance Male 11 93.5455 3.01210 .90818

Female 14 83.3571 3.10353 .82945

105
 Referring to the Independent Samples Test table again we can see the results of

the t-test. The t-value is 8.253, there 23 degrees of freedom and the (two-tailed)

significance level is 0.000. The difference between two groups’ means is significant in

this case because 0.000 is less than to the assigned level of significance (0.05).

The mean teaching performance were 93.55 (SD = 3.01) for males and 83.36

(SD = 3.10) for females. The teaching performance of males and females were

compared using and independent samples t-test. There was no significant difference

between the teaching performance of males and females t(23) = 8.253, p-value = 0.000.

106
SIGNIFICANCE OF THE DIFFERENCE BETWEEN TWO MEANS FOR DEPENDENT
SAMPLES

In the conduct of research, one may use two related samples, that is, one may

“match” or otherwise relate the two samples studied. This matching may be achieved by

using each subject as his own control (as in pre-post tests design) or by matching

subjects and then randomly assigning one member of each pair to one of the two

conditions, and his matched partner to the other condition. A procedure for testing the

significance between the mean difference of two related samples may be applied

without computing the correlation coefficient between them. This method is sometimes

called the difference method.

Let D be the difference between any pair, x 1-x2 = D for any pair. Summing over N

pairs yield ∑ X 1 −∑ x 2 =∑ D . Dividing by N we obtain x́ 1− x́ 2= D́ . Since the mean

difference is the difference between the two means, we may test for the significance of

the difference between two means by testing whether or not D́ is significantly different

from zero. Here, we treat D as a variable.

The unbiased estimate of the variance of D is:

S2D=∑ ¿ ¿¿

2 S2D
S =
D́
N

To test whether D́ is significantly different from 0, we use the t-ratio of:

D́
t=
S D́
An equivalent formula is:

107
 t=
∑D
2 2

Example:
√ N ∑ D −( ∑ D )
N −1

The following are the pre and pos-test scores of 12 students on a test which

determines whether learning has taken place as a result of a new teaching strategy.

Test the significance between the two means at the 5% level.

Subject No. Pre-Test Post-Test

1 18 22
2 14 11
3 9 10
4 17 15
5 7 11
6 16 19
7 12 14
8 11 13
9 11 13
10 17 20
11 15 16
12 8 10
H 0 :There is no significant difference between the mean of the pre-test and post-test.
H 1 : There is a significant difference between the mean of the pre-test and post-test.
Subject No. Pre-Test Post-Test D D2
1 18 22 -4 16
2 14 11 3 9
3 9 10 -1 1
4 14 15 -1 1
5 7 11 -4 16
6 16 19 -3 9
7 12 14 -2 4
8 11 13 -2 4
9 11 13 -2 4
10 17 20 -3 9
11 15 16 -1 1
12 8 10 -2 4
2
∑ D=−22 ∑ D =78

108
 t=
∑D
2
Use 2

√ N ∑ D −( ∑ D )
N −1
−22
t= =−2.95
12 (78 )−(−22)2
√ 12−1
−22
t=
936−484
√ 11
−22
t=
√ 41.09
−22
t= =3.43
6.41
Tabular value of t at 11 df, 5% = 2.201
Observed Value of t = -3.43

Conclusion:

Since observed value of t < tabular value at 5% level of significance, then the null

hypothesis is not rejected. This means that the post-test mean is not significantly

different from the pre-test mean. It further means method did not increase achievement
exposure to the new teaching method.

SIGNIFICANCE OF THE DIFFERENCE BETWEEN TWO MEANS FOR DEPENDENT

SAMPLES IN SPPS
Select Analyze, Compare Means, Paired-Samples T-test… from the Data

Editor window

109
 From the Paired Sample t-test window, select pre test as variable 1 and post test
as variable 2. Then, click the “ok” button.

110
 The following table will show the result.

Paired Samples Statistics

Mean N Std. Deviation Std. Error
Mean
Pre_test 12.6667 12 3.57601 1.03231
Pair 1
Post_test 14.5000 12 4.03395 1.16450

Paired Samples Correlations

N Correlation Sig.

Pre_test &
Pair 1 12 .889 .000
Post_test
Paired Samples Test

Paired Differences t df Sig. (2-

tailed)

Mean Std. Std. Error 95% Confidence Interval of

Deviation Mean the Difference

111
 Lower Upper

Pre_test - -
Pair 1 1.85047 .53418 -3.00907 -.65760 -3.432 11 .006
Post_test 1.83333

MORE SPSS APPLICATION: PAIRED SAMPLES t-TEST (DEPENDENT)

A researcher is interested whether a training course increases or decreases the
teaching performance of the teachers who attended the training. The researcher
randomly selected 25 teachers and measured their teaching performance before and
after they attended the training courses. The data are shown in Table 3.

Table 3. The Teaching Performance of 25 Teachers

Before and After a Training Program

Case Teaching Performance Teaching Performance

Before Training After Training
1 85 95
2 84 98
3 86 97
4 87 92
5 89 96
6 82 93
7 80 94
8 84 95
9 86 90
10 82 82
11 89 97
12 87 98
13 82 95
14 81 95
15 86 92
16 89 91
17 89 94
18 84 95
19 85 96
20 88 97
21 81 99
22 80 98
23 86 99
24 87 91
25 89 98

112
Step 1. Enter your data in the SPSS Data Editor.

Figure 3.1 Paired samples t-test data entered in SPSS Data Editor

Step 2. Run Paired Samples t-test.

 Click on Analyze.
 Click on Compare Means.
 Click on Paired Samples T Test.
 A dialog box will appear. Click on both the variables you want to include in the
analysis, and then click on the small black arrow to the left of the white box
headed Paired Variables. The variables will move into the white box as in shown
in Figure 3.2.
 Click OK.

113
Figure 3.2 Entering variables in a paired samples t-test.

The output will be as follows

Paired Samples Statistics

Mean N Std. Std. Error
Deviation Mean
before_TeachinPerfromance 85.1200 25 3.00444 .60089

Pair 1
after_TeachingPerformance 94.6800 25 3.70495 .74099

Paired Samples Test

Paired Differences t df

114
 Mean Std. Std. 95% Confidence Sig.
Deviati Error Interval of the (2-
on Mean Difference tailed)
Lower Upper
before_Teachi
nPerfromance
- 4.6195 . - -
Pair 1 - -10.347 24 .000
9.56000 2 92390 11.46685 7.65315
after_Teachin
gPerformance

The mean Teaching Performance was 85.12 (SD = 3.00) before the training

course and 94.68 (SD = 3.70) after the training course. The difference between the

mean level of Teaching Performance before and after the training course was examined

with a paired samples t-test. Teaching Performance was significantly higher after the

training course than before it t(24) = 10.347, p-value < 0.05.

Analysis of Variance (ANOVA)

Analysis of variance commonly abbreviated as ANOVA is an interval model used

to compare the means of three or more groups. For example, if you want to compare

the mean achievement test scores in science and math in the National Achievement

Test (NAT) of students in three (3) schools in Metro Manila, you use ANOVA.

Assumptions in the use of ANOVA.

 The variables/s must be normally distributed.

115
  Random sampling method was used in the selection of the samples

 The variances of the subgroups must be homogeneous

 The samples that constitute the subgroups have been drawn independently.

A typical ANOVA Table

Source of Degrees of Sum of Mean F-ratio ρ

variation Freedom Squares Square
Between
groups
Within
groups
Total

ANOVA relies on the F-ratio to test the hypothesis that the two variances are

equal; that is, the subgroups are from the same population. “Between groups” refer to

the variation between each group mean and the grand or overall mean. “Within groups”

refer to the variation between each subject and the subject’s group mean.

Between groupmean square

F=
Within group mean square

116
Computational Technique for Single Classification

Sample Data

Group
I Group A B Group C A B C
1 15 16 8 225 256 64
2 12 15 6 144 225 36
3 16 17 16 256 289 256
4 10 16 6 100 256 36
5 8 10 8 64 100 64
6 14 15 14 196 225 196
7 12 8 16 144 64 256
Total 87 97 10.57 1129 1415 908
Group  
Means 12.43 13.86 10.57 258/21 = 12.29

   
           

TOTAL SUM OF SQUARES

Variation of all the observed values from the grand/overall mean:

SST =X 2−( X )2/N

=3452-(258)2/ 21 = 282.28

BETWEEN SUM OF SQUARESS

Variation of the group means fro the grand/overall mean

SSB=∑ (X −X g)2 n1 ,n 1=group sample ¿ ¿

This can be written as:

SSB=∑ n1 (X 1)2 −( ∑ X ) /n
2
87 2 972 74 2 ( 258 )
¿ (7
+
7
+
7
−
21 ) =38

WITHIN SUM OF SQUARES

Sum of the variations within the groups

SSW =∑ SS G1

117
 1. SS G A= ∑ X 2A −(∑ X A ) /n A
2 2 2 2 ( 87 )2
2 2 2
¿ ( 15 +12 +16 +10 +8 +14 +12 ) −
7
¿ ( 225+144+ 256+100+64+ 196+144 )−1081.29
¿ 1129−1081.29=47.71

2. SS G B =∑ X B2−
( XB )
nB
( 16 2+152 +17 2+16 2+ 102+15 2+ 82 )−¿ ¿
( 256+225+ 289+ 256+100+225+64 )−1344.14
¿ 1415−1344.14=70.86

3. SSG C = ∑ X 2C−¿ ¿ ¿
¿ 82 +6 2+16 2+ 62 +82 +142 +16 2 ¿−¿ ¿
¿ ( 64+ 36+256+36+ 64+196+256 )−782.29
¿ 908−782.29
¿ 125.71

SSW =47.71+ 70.86+125.71

¿ 244.28

RELATIONS OF THE SUM OF SQUARES

SST= SSB+SSW

SSW = SST – SSB

= 282.28 – 38 – 244.28

ANOVA Table

Source of Variation Df Sum of squares Mean square

Between groups 2 38 19
Within groups 18 244.28 13.57
Total 20 282.28

Calculating the F-ratio

Between groupmean square

F=
Within group mean square

118
In this example, the computed F-ratio is given by:

F = 19/13.57 = 1.4 (with 2 and 18 d.f)

On the other hand, the tabulated value for F with 2,18, d.f. is 3.55 for α=0.05.

Interpretation of the result:

Since the computed F-ratio of 1.4 is less than F0.05(3.55), we can accept the null hypothesis
and say that the variability between subgroup is not significant to say that they belong to the same
population.

Analysis of Variance (ANOVA) in SPSS

To run analysis of variance in SPSS, select Analyze, Compare Means, One way ANOVA... from the data
editor window.

119
 Then specify scores in dependent variable box and student in the factor box. Then click “ok”
button.

The following table will show the ANOVA result.

ANOVA
Scores

Sum of Squares df Mean Square F Sig.

Between Groups 38.000 2 19.000 1.400 .272

Within Groups 244.286 18 13.571
Total 282.286 20

MORE SPSS APPLICATION ONE-WAY ANALYSIS OF VARIANCE (ANOVA)

120
 In this example a researcher is interested whether teachers’ “employment status” differ in

relation to teaching performance. To investigate this the researcher selected 8 teachers in each type of

employment status and measured their teaching performance. The data are shown in Table 4. Teachers

are coded 1 if they are part-timer, 2 if they are temporary and 3 if they are permanent.

Table 4. The teaching Performance of 24 Teachers in

Three Types of Employment Status

Case Employment Status Teaching Performance

1 85 1
2 84 2
3 86 3
4 87 1
5 89 3
6 82 2
7 80 1
8 84 3
9 86 2
10 82 1
11 89 3
12 87 2
13 82 1
14 81 3
15 86 1
16 89 2
17 89 1
18 84 3
19 85 2
20 88 1
21 81 3
22 80 3
23 86 2
24 87 2

121
Step 1. Enter your data in the SPSS Data Editor.

Figure 4.1 Independent Samples ANOVA data entered in SPSS Data Editor

Step 2. Run Paired Samples t-test

 Click on Analyze.
 Click on Compare Means
 Click on One-Way ANOVA
 A dialog box will appear. Click on the dependent variable. It will be highlighted in black. Click on
the black triangle to the left of the box headed Dependent List (Teaching Performance). The
dependent variable will be moved across to this box.
 Click on the independent variable. It will be highlighted in black. Click on black triangle to the left
of the box headed Factor (Employment Status).
The dependent variable will be moved across to this box (see Figure 4.2).

122
  Click on Post Hoc...A dialog box will appear.
 Click on the white box next to Tukey. A tick will appear in the box (see Figure 4.3).
 Click on Continue.
 Click on Options...
 Click in the white boxes next to Descriptive (see figure 4.4)
 Click on Continue.
 Click OK.

Figure 4.2 Entering the independent (Factor) and dependent variables when carrying out in independent
samples ONE-WAY ANOVA.

123
Figure 4.3 Specifying Post Hoc Multiple Comparisons when carrying out an independent samples ONE
WAY ANOVA

124
Figure 4.4 Entering the options when carrying out (Descriptive) an independent samples ONE WAY
ANOVA

Descriptives
Teaching_Performance

95% Confidence Interval

for Mean
Std. Deviation Lower Upper
N Mean Std. Error Bound Bound Min Max

Part-timer 8 84.8750 3.22656 1.14076 82.1775 87.5725 80.00 89.00

Temporary 8 85.7500 2.12132 .75000 83.9765 87.5235 82.00 89.00
Permanent 8 84.2500 3.53553 1.25000 81.2942 87.2058 80.00 89.00
Total 24 84.9583 2.95590 .60337 83.7102 86.2065 80.00 89.00

ANOVA

125
Teaching_Performance

Sum of Squares df Mean Square F Sig.

Between Groups 9.083 2 4.542 .497 .615

Within Groups 191.875 21 9.137
Total 200.958 23

Multiple Comparisons
Dependent Variable: Teaching_Performance
Tukey HSD

(I) Employment_Status (J) Mean 95% Confidence Interval

Employme Difference (I- Lower Bound Upper Bound
nt_Status J) Std. Error Sig.

Temporary -.87500 1.51137 .833 -4.6845 2.9345

Part-timer
Permanent .62500 1.51137 .910 -3.1845 4.4345
Part-timer .87500 1.51137 .833 -2.9345 4.6845
Temporary
Permanent 1.50000 1.51137 .590 -2.3095 5.3095
Part-timer -.62500 1.51137 .910 -4.4345 3.1845
Permanent
Temporary -1.50000 1.51137 .590 -5.3095 2.3095

All p-value is greater than to the assigned level of significance which is 0.05, therefore
allpairs are not significantly different.

The mean levels of teaching performance were compared between teachers with
employment status of part-timer, temporary and permanent. The mean levels of teaching performance
were 84.8750 (SD=3.22656) for part-timer, 87.7500 (SD=2.12132) for temporary and 84.2500
(SD=3.53553) in permanent. An independent samples one-way analysis of variance showed that there is
no significant difference between the teaching performance across the types of employment status
F(2,21)=0.497, p>0.05.

Reading Assignment:

 E- Journals/E-Books:
 PUP website: infotrac.galegroup.com/itweb/pup
 Password: powersearch

Exercises/ Written Assignments:

1. Ten teachers who applied for the Licensure Examination for Teachers (LET) were given a test to
measure their readiness. They were subjected to refresher courses after which a second test
was given to determine if the course affects their readiness. The results were given below.

126
 Teachers T1 T2
1 125 157
2 89 131
3 135 155
4 119 162
5 190 245
6 135 148
7 186 225
8 147 170
9 180 195
10 145 135

Is the refresher course effective?

2. Ten students are paired in terms of age and year level. One member of each pair taught Physics
in English and the other member using Filipino. At the end of the course, a proficiency test was
administered.
The data are tabulated below.

Student 1 English Filipino

1 70 82
2 80 80
3 70 81
4 69 70
5 80 86
6 90 55
7 87 82
8 75 35
9 82 78
10 80 85
Is there a significant difference?

3. The following represent the test scores of three (3) groups of respondents

A B C
12 6 7
11 8 4
10 6 10
9 10 9
14 8 12
12 6 4
13 7 3
14 9 6
11 12 5

127
 10 13 6

3.1 Apply analysis of variance to see if the groups differed significantly in their performance. Use
alpha =.05

References/ Bibliography

Downie, N.M. & Heath, Robert W. (1984). Basic Statistical Methods (Fifth Edition).

Edwards, Allen L. Statistical Analysis. (1988). New York, Holt, Rinehard and Winston.

Ferguson, George A. (1989). Statistical Analysis in Psychology and Education. New York:

McGraw-Hill.

Field, Maurice H. (1997). Statistics for Graduate Research Projects. Tennessee.

Guilford, J.P. (1985). Fundamental Statistics in Psychology and Education. New York:

McGraw-Hill.

Kumar, R. (1996). Research Methodology, pp. 1-15. Busy Book Distributors.

Lwwsy, P.D. and Ormrod, Jean E. (2005). Practical Research Planning and Design, 8th edition,

pp. 1-6, Prentice Hall.

Naval, Victoria C. et al. (1988). General Statistics. Conanan Publishing.

Ostle, Bernard B. (1988). Statistics in Research: Basic Concepts and Techniques for Research

Workers. Ames: lowa State University Press.

Van Dalen, Deobold B. (1982). Understanding Educational Research. New York: McGraw-Hill.

128