CHAPTER IV METHODOLOGY 4.1 Variables of the Study 4.2 Tools Used for Measurement 4.3 Sample Used for the Study 4.4 Data Collection Procedure 4.5 Scoring and Consolidation of Data 4.6 Statistical Techniques Employed 4.7 Other Details Relating to the Desigr
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CHAPTER IV
METHODOLOGY
4.1 Variables of the Study
4.2 Tools Used for Measurement
4.3 Sample Used for the Study
4.4 Data Collection Procedure
4.5 Scoring and Consolidation of Data
4.6 Statistical Techniques Employed
4.7 Other Details Relating to the Desigr
METHODOLOGY
In order to study the personality correlates of Process Outcomes in
Basic Science of Upper Primary School Children, the following procedures are
adopted. The details of these procedures are briefly described below under
the following heads:
1. Variables of the study
11. Tools used for measurement
111. Sample used for the study
IV. Data collection procedure
V. Scoring and consolidation of data
VI. Statistical techniques employed.
4 .1 VARIABLES OF THE STUDY
The study has been designed mainly with Process Outcomes in Basic
Science as the major dependent variable and a set of personality variables as
the independent variables. The details relating to the variables of the study
are presented below:
4.1.1 Dependent Variable
Major dependent variable: Process Outcomes in Basic Science, which is
composed of the following four minor dependent variables:
Process I - Classifying : Imposing order and collecting objects and events.
Process I1 - Reasoning : Making decisions and judgements by thinking
clearly and logically.
Process Ill - Inferring : Making an explanation of an observation.
Process IV - Predicting : Making specific forecasts of what a future
obseivation will be.
4.1.2 Independent Variables
Five personality variables, treated as independent variables are given
below:
( a ) personal adjustment;
(b) social adjustment;
(c) examination anxiety;
(d) achievement motivation;
(e) science interest.
4.2 TOOLS USED FOR MEASUREMENT
The psychometric and other details relating to the tools used in -the
study are described below:
4.2.1 Test of Process Outcomes in Basic Science
The Test of Process Outcomes in Basic Science was developed and
standardised by the investigator in collaboration with her supervising teacher.
Process outcomes were operationalised on the basis of the theoretical models
developed by (a) the American Association for the Advancement of Science
(AAAS. 1968) and (b) Scientific Research Processes (Kerlinger, 1973). On the
basis of the classification of skills presented here and in consultation with
experts in science education, the investigator has finalised the Test of Process
Outcomes in Basic Science (for Upper Primary School Children).
4.2.1.1 Procedure
The procedures used for developing the test are summarised below.
4.2.1.1 .I Item heparation
Since the purpose of the present test was to measure Process Outcomes
in Basic Science of Upper Primary School Children, the items prepared were
from the areas of the syllabus of Basic Science of Upper Primary Schools of
the Kerala State. Twenty-five items each were pooled initially for the four
subcategories.
All the items were prepared by the investigator herself. These items
were subjected to scrutiny and criticism by a team of experts in science
education.
The draft test was pre-tried out for twelve students randomly selected
from Government Teachers' Training Institute, Ettumanoor in order to
understand the clarity of wording and directions of the test.
4.2.1.1.2 heparation of Draft Test
On the basis of the pre-try out, fifteen i t ~ m s each were selected for
process 11 (reasoning), process 111 (inferring), process IV (predicting) and
sixteen items for process I (classifying).
The two-way blueprint of the draft test showing the number of items
allotted to each process and content area of the Basic Science syllabus was
prepared. It is given in Appendix A.
The draft form of the test contained 61 multiple choice items with four
alternatives each. The content area in the Upper Primary Basic Science which
fall into seventeen chapters were analysed to identify the processes to be
tested. The content area of the Basic Science Curriculum and the number of
items set under each are given in the Table 2.
TABLE 2
Content Area of the Basic Science Syllabus and the Number of Items from Each in the Draft Test
Content area in the svllabus No. of items set
Growth and response to stimuli 4
Living things adapt themselves
Communicable diseases
Community sanitation
Food and deficiency diseases
Variety in organisms 4
Structure and functions of the living body 6
Atoms and molecules
Separation of substances
Changes around us
Motion and force
Work and energy
Shadows and eclipses
Universe
Air around us
Balance in nature 1
Life processes 6
Total 61
All the items were such as would measure a single defined process
outcome in Basic Science at the Upper Primary level. A minimum of fifteen
and a maximum of sixteen items were prepared for each process categoy.
The details regarding the number of items in each process and the item
numbers corresponding to each process in the draft test are presented in
Table 3.
TABLE 3
Number of Items in the Draft Test Classified into Process Categories
Process No. of items in the Item nos. in the catego y process catego y draft test
Classifying 16 1-16
Reasoning 15 17-31
Inferring 15 32-46
Predicting 15 47-61
Total 61 -
The draft items prepared were exposed to expert criticism. Specialists
in Science Education belonging to important Teacher Education Colleges,
Colleges of Science and University Departments of Education were consulted.
The draft test was criticised by the specialists. Modifications were made on the
basis of the suggestions made by the specialists and the final draft was
prepared.
The Test of Process Outcomes in Basic Science (draft form) containing
61 items with directions to the candidates was printed in a booklet form of
eight pages (vide Appendix B). Response sheet of the draft test is given as
Appendix C.
4.2.1.1.3 Administering the Drat? Test
The draft test was administered to a representative sample of 378
pupils studying in standard VII of thirteen schools of Kottayam District. The
testing was done during the academic year 1999-2000. The draft test was
administered by the investigator herself. Details of the schools seletted for the
draft test is given in Table 4. The test was administered without time limit.
Most of the students completed the test in about one hour. All the students
could complete the test in one hour and ten minutes.
TABLE 4
Details Regarding the Schools Selected for Draft Test
SI. No Narnc. of the school School Type of Type of No. of No, of Total location school management boys girls
The scores of all tests of the selected subjects and other data relating to
them were tabulated on a consolidated data sheet. For each variable one
column was provided. Hence five columns for process variables-classifying,
reasoning, inferring, predicting, process total, five columns for five personality
variables, one for sex, one for school location and the last column for type of
school management were set apart. The analysis has been based on the data
relating to the 600 subjects obtained as explained above.
4.6 STATISTICAL TECHNIQUES EMPLOYED
The objectives of the study and the hypotheses formulated for the study
suggested the use of the following statistical techniques for data analysis.
4.6.1 Two-tailed Test of Significance of the Difference Between Means for Large Independent Groups
For this to work out the t-values (also called the critical ratios) given by
the formula
M, = Mean test score for the first group
M, = Mean test score for the second group
and
SE,M,-M21 = The standard error of the difference between means M, and M,
SE,MI-M21 was calculated by the formula
where SEMI and SEM2 were the standard error of the mean scores MI and M,
respectively.
If N, and N, are the size of the samples under comparison and o, and
a,, their standard deviations,
The obtained t-value (critical ratio) was then treated as belonging to a
normal distribution. If the obtained t-value falls between -1.96 and +1.96,
the difference between means was treated as being not significant at 0.05
level. In this case. the difference was treated as the same as a zero difference.
If the t-value falls outside the interval 0.05 level, this means that the difference
was real and was greater than zero.
If the obtained t-value falls outside the interval +2.58. the difference
was treated as significant at 0.01 level, otherwise the difference was treated as
not significant at 0.01 level. A significant difference between means imply that
the difference was real and was different from zero. A non-significant
difference indicated that the difference between the means was not real and
indicated difference IS to be attributed to sampling errors.
4.6.2 Person's Product-moment Coefficient of Correlation
When a set of 'n' pair of scores for two continuous variables x and y are
given in the form of (x,. y,), (x,, y,), (x,, y,) . . . (x,, y,) the correlation between
the variables x and y represented as r,, is given by the following formula
where Cx = the sum of all x scores in the data
xy = the sum of all y scores in the data
2x2 = the sum of the squares of all the x scores in the data
Cy' = the sum of the squares of all the y scores in the data
Xxy = the sum of the produds of all the paired x and y values of the
data.
The obtained correlation coefficients were interpreted by means of the '
following approaches.
4.6.2.1 Test of Significance of the Correlations
This was done by checking whether the obtained correlation exceeded 1 1 - x 1.96 or - x 2.58 for significance at 0.05 level and 0.01 level fi f i respectively.
4.6.2.2 The 0.01 Confidence interval of 'r 's
The limits of the 0.01 confidence interval was estimated using the
formula ( r f 2.58 SEr), in which SEr, the standard error of r, was obtained
using the formula
I ,.' SEr = --
G i
r, being the obtained coefficient of correlation.
4.6.2.3 Verbal Descriptions
For interpreting the values of r, verbally, Garrett's (1981) classification
was used.
i.e. 'r' from 0.00 to k0.20 denotes indifferent or negligible relationship;
'r' from 0.20 to i0.40 denotes low correlation present but slight;
'r' from 0.40 to k0.70 denotes substantial or marked relationship;
'r' from 0.70 to k1.00 denotes high to very high relationship.
4.6.2.4 Percentage Variance (Fox, 1969)
This was estimated by finding out ? x 100, r, being the obtained
correlation coefficient.
4.6.3 Multiple Regression Equation
In order to examine the relationship between Process Outcomes .in
Basic Science and various independent variables a multiple regression
equation was formulated. The regression equation is
where x, - score of Personal Adjustment
x, - score of Social Adjustment
x, - score of Examination Anxiety
x, - score of Achievement Motivation
xi - score of Science Interest
Y - score of Process Outcomes in Basic Science
E - random disturbance term
a - is a constant
b, - regression coefficient of Personal Adjustment
b, - regression coefficient of Social Adjustment
b, - regression coefficient of Examination Anxiety
b, - regression coefficient of Achievement Motivation
b, - regression coefficient of Science Interest
a, b,, b,, b,, b, and b, are estimated using Ordinary Least Squares (OLS)
method.
4.7 OTHER DETAILS RELATING TO THE DESIGN
Procedure used for categorising the whole sample into groups based on
differing levels of Process Outcomes in Basic Science, viz. high achievers
(HA), average achievers (AA) and low achievers (IA).
The total sample was divided into three goups (HA, AA and LA) based
on the scores obtained for Process Outcomes in Basic Science, by the whole
sample. The following procedure was used for the division. The score of the
total sample of 600 subjects in Process Outcomes in Basic Science was use I
for calculating mean and standard deviation of the score distribution.
Assuming that M is the mean score and o. the standard deviation of the
600 scores of Process Outcomes in Basic Science, the groups were labelled as
below.
A subject whose score on Process Outcomes in Basic Science fell
between M k o was classified as an average-achiever (AA). A subject whose
score was below M - o was classified as a low-achiever (LA). A subject whose
score was above M + o was classified as a high-achiever (HA). In the present
study Mean (M) of process outcome scores in Basic Science for the total group
was 25.72 and standard deviation (o) was 6.78. Therefore, subjects whose
scores as Process Outcomes in Basic Science were 33 and above (rounded
value of M + o) were labelled as High Achievers (HA); those subjects whose
scores were less than 19 (rounded value of M - o) were classified as Low
Achievers (LA) and all others were treated as Average Achievers (AA).
The number of subjects so obtained in each of the three groups are
presented in Table 11.
TABLE 11
Number of Subjects Falling within the Three Groups Based on Process Outcomes in
Basic Science (HA-, AA- and LA-)
Process Outcomes Group Total No. of Subjects
High Achievers (HA) 103
Average Achievers (AA) 413
Low Achievers (LA) 84
Details of the analysis are presented in the next chapter.