1. The document discusses various methods for developing and evaluating knowledge tests, including item collection, screening, analysis of item difficulty, discrimination, and point biserial correlation.
2. Reliability of the knowledge test is determined using split-half methods like the Spearman-Brown formula.
3. Knowledge scores are calculated and can be categorized into different levels to interpret respondents' knowledge.
Development of knowledge test and some other tests used in social science research
1. Authors:
DR. SUKANYA BARUA, DR. R.N PADARIA and
DR. PREMLATA SINGH
ICAR- INDIAN AGRICULTURAL RESEARCH INSTITUTE
2. A test is a set of questions, each of which has
a correct answer, to which the people
respond
A test is an organized succession of stimuli,
designed to measure quantitatively some
mental process, trait or charecteristic
3. - Systematic procedure
- Comparing behaviour of two or more person
- Time (At a time/different time)
- Among persons/ Between same person
4. Knowledge as a body of understood
information possessed by an individual or a
culture
It is a Components of behaviour
Influences covert and overt behaviour of a
person
5. Knowledge test score used as a variable to test
its relationship with other variables
Steps in Knowledge test:
1. Item collection
Initial selection of items should promote
thinking rather than only memorisation
Should differentiate between well informed
respondents from poorly informed one
6. in consultation with subject matter experts
Purpose of study
Idea about education level of respondents
Idea about existing level of knowledge of
respondents
The items should be in objective form and
yes/no, true/false, multiple choice etc.
7. 2. Screening out of statements after consultation
with experts
3. Item analysis:
Based on,
i. Item Difficulty
ii. Item discrimination
iii. Point biserial correlation co-efficient
8. The difficulty index indicates the extent to
which an item is difficult
An item should not be too easy that all persons can
pass it, nor should be too difficult that none can
pass it
9. P= NC×100
N
P = Difficulty index,
NC = Number of respondents who correctly
answered
N= Total number of respondents
Difficulty index value should ranges from 30-80
10. The function of discrimination index is to
find out whether an item really discriminates
from a well informed respondent to a poorly
informed respondent
Score of total respondents should be rank in
descending order
11. Respondents should divide into six equal
groups i.e. G1, G2, G3, G4, G5, G6
The middle two groups i.e. G3 and G4 were
eliminated and the remaining four terminal
groups i.e., high score groups (G1 and G2)
and low score groups (G5 & G6) remained for
further analysis.
12. DI= (S1+S2)-(S5+S6)
N/3
DI =Discrimination index, S1, S2, S5, and S6
are the frequencies of correct answers in the
group of G1, G2, G5, and G6, respectively. N
is the total number of respondents in sample
selected for item analysis
13. For establishing the internal consistency of
each item, point biserial correlation
coefficient (rpbi) was estimated
Point biserial correlation co-efficient:
14. rpbi = Point-biserial correlation coefficient, Xp =
Mean score of continuous variable in successful
group on dichotomous variable, Xq= Mean score
of continuous variable in unsuccessful group on
dichotomous variable, St = Standard deviation of
continuous variable for total groups, p =
Proportion of respondents who answered
correctly on dichotomous variable and q = 1-p or
the second group.
15. rpbi value should be equal to or more than
0.236 was considered for the selection in the
final knowledge test
16. The reliability of the knowledge test can be
determined by the following Split-Half
method or Spearman-Brown formula
rtt = reliability coefficient of the test
rhh = the correlation between two halves of
the test.
17. The standardized knowledge test should be
administer and results will be interpreted by
giving ‘1’ against correct response and ‘0’
against incorrect response respectively
The sum of all the scores on each item will
be the total knowledge score for the
individual respondent
18. Knowledge test can be descriptive type or
mixed type (both objective and descriptive)
In case of descriptive type knowledge test
marks have to be given by experts or can be
set some criterion for assigning values.
19. Knowledge score obtained by respondent×100
Maximum obtainable knowledge score
Knowledge score can be categorized into different levels like
Low, Medium, High
Or,
Very low, low, Medium, High, Very High
Or,
Any other
20. Intelligence tests:
It refers to general set of mental abilities
To achieve means to accomplish, to attain a
desired end
1. Individual intelligence test
- Binet-Simon scale was the first one
2. Group intelligence test
- Army alpha test and army beta test
21. It can be further divided into,
1. General Achievement test: It attempt to
measure the general educational
achievement in several years of
academic curriculum
Eg: The Iowa test of educational
development, California achievement test, Iowa
test of basic skills
2. Special Achievement tests meant for
measuring the achievements of pupils in some
selected areas. It may be grouped into the
diagnostic tests and the standardized end of
course examinations
22. Aptitude is a person’s ability, acquired or innate,
to learn or develop knowledge/ skill in some
specific area
1. Multiple aptitude test
2. Special aptitude test
Multiple aptitude test are those which intend to
measure several aptitudes, each by an
independent sub test
Special Aptitude test are those which intend to
measure only one aptitude
23. Motor dexterity test
Artistic aptitude test
Musical aptitude test
Test of creativity
Test of mechanical aptitudes
Clerical aptitudes etc.
24. Yadav et al. 2013. Development of a Test for
Measuring the Knowledge Level of Women
Farmers in Vegetable Cultivation. Journal of
Human Ecology, 41(2): 113-117.
Kerlinger, F. N. 1964. Foundations of
Behavioural Research, Holt, Rinehart and
Winston Inc. New York.
Ray and Mondal. 2011. Research Methods in
Social Sciences and Extension Education,
Kalyani Publishers, pp.179.