Sei sulla pagina 1di 6

Running Head: RELIABILITY AND VALIDITY

Reliability and Validity in Assessment in AG Education

Name.

Institution.

Date.
RELIABILITY AND VALIDITY 2

An instrument's validity is the measurement that the tool tests what it wants to measure

whereas reliability refers to the consistency of measure (Mohajan, 2017). This research focuses

on two groups; a class and FAA assessment whose outcomes were as follows.

Score at class Score at FAA


50% 60%
46% 45%
53% 78%
52% 54%
67% 56%
59% 55%
53% 52%
45% 47%

Cronbach Alpha for a sample of 8 students in the class was 0.896, indicating high reliability as shown in
the table below;

Reliability Statistics

Cronbach's Cronbach's N of Items


Alpha Alpha Based on
Standardized
Items

.896 .416 2
RELIABILITY AND VALIDITY 3

Reliability
It is evident that the sample uses test-to-test reliability as various outcomes were repeated

accordingly. Using the two test-scores is done by graphing and using inter-correlation as shown

in the graph below;

Correlations
Class FAA
Pearson Correlation 1 .529
Class Sig. (2-tailed) .263
N 8 8
Pearson Correlation .529 1
FAA Sig. (2-tailed) .263
N 8 8

From the graph it is clearly shown that elements were scattered. There was relatively low

correlation-ship between the two test scores. Pearson’s Correlation indicated was at 0.529. In

general a test-re-test that indicates +.55 shows relatively high correlation hence good reliability.
RELIABILITY AND VALIDITY 4

However, from the data, there was a significant correlation relatively to 0.55. The two tests had a

significant correlation hence the observations had no changes in a substantial manner.

As with the reliability of test-retests, internal consistency can only be measured by data

collection and analysis (Sabour, 2017). The score computed for each items indicated a positive

correlation of 0.526. This was a relatively high correlation. In general, from Pearson’s

Correlation, correlation of +0.45 or greater is considered as a good internal consistency. This is

applicable from the data analyzed.

Content Validity

Content validity is an estimation of how well a set of element scales suits the applicable

content domain of the construct it is attempting to calculate (Sabour, 2017). It applies to the

degree to which the items on a test are equally reflective of the entire area that the test aims to

assess. It is a qualitative form of validity where the definition domain is made clear and the

analyst assesses whether the measures reflect the domain in full. The sample above relied on the

knowledge of students familiar with what was expected for measurements.

For example, from the above analysis we could conceptually describe a study anxiety as

including both sympathetic stimulation of the nervous system or a negative attitude resulting to

an aggregate performance of the both class and contest work. By this definition a person have

high marks is pre-assumed that the student feels positive in learning and actually reads frequently

and learn. Therefore, in order to have a good content validity, then the measurement of the

student’s high marks should reflect the corresponding aspects.


RELIABILITY AND VALIDITY 5

Face Validity

Face validity is a measure reasonably for a validity construct on its face (Sabour, 2017).

A good example is that most people expecting a test on self-esteem to include elements about

whether they see themselves as deserving or if they believe they have good qualities. A

questionnaire containing such items would have a good face validity. However in our case the

students can opt to choose whether to measure their IQ levels, high or low or as well measure the

different levels of discipline as a result of the scores.

Face validity is at best a very poor kind of proof that a system of measuring tests what it

is meant to be. That's because it's based on people's intuitions about human behavior, which are

always incorrect. Judgmental cases could be wrong in this analysis. An example, is that a third

party could suggest that having a score of 55% and above in a class assessment means that the

resulting FAA contest should be positive also as the level of intelligence is high. However, there

may be general assumptions that could result to a low or high performance, for example, external

factors, health issues, and many more. In this case, it is not the direct responses of the

participants to these questions that are of concern but rather how the frequency of the responses

of the participants to a set of questions fits that of people that appear to conceal their rage

(Sabour, 2017).
RELIABILITY AND VALIDITY 6

References.

Mohajan, H. K. (2017). Two criteria for good measurements in research: Validity and reliability.

Annals of Spiru Haret University. Economic Series, 17(4), 59-82.

Sabour, S. (2017). Contraceptive knowledge assessment: methodological issue on reliability

analysis. Contraception, 96(4), 257.

Potrebbero piacerti anche