Social Science Research


The conclusions and propositions of any social science research must be reliable and valid, reliability and validity are related but are different in their perception, this paper focuses on the definitions and meaning of validity and reliability and also shows the relationships that exists between both concepts. In any research we can have valid and reliable results, reliable but invalid results, valid and non reliable results and finally invalid and unreliable results.


Reliability means that the results of a study are of quality in measuring relationships between variables, when measures are reliable it means that if another study was undertaken the same results will be obtained, this means that the measure of the study is consistent and repeatable if other studies of the same type are undertaken.

There are various reliability measures which include observer reliability, test reliability, parallel reliability and consistent reliability:

Observer reliability:

This involves the efforts to establish where there observations are consistent with the others, example a study involving the interview of two individuals then there is a need to compare the two observations and establish whether the results from both observations are consistent and therefore reliable.

Social Science Research

Test reliability:

In test reliability we try to establish the correlation between studies undertaken in two time periods, it is evident from research that given the same sample and undertake the study tafter a short time period then the higher the correlation but if we took the same sample and the have a longer time period then the lower the correlation.

Parallel reliability:

This involves testing reliability using many set of questions that tend to yield same results, in a study questions can be split into two where the second part tends to establish the same results as the first part, this way the similarities will help establish whether the results are consistent and reliable.

Consistency reliability:

This involves establishing a single measure of consistence on a group of participants at the same time and try to establish whether the results are consistent, this helps in determining whether a study undertaken on the group will be consistent.

All the above methods are used in establishing whether the results of a study are reliable or not, however the methods have their strength and weakness but they are important in determining reliability.

Social Science Research


The validity can be defined as the best approximation of truth of an inference, however the validity of a study does not depend on the use of good study design or even use of good samples in the study. Validity measure is divided into four including conclusion validity, internal validity, construction validity and finally external validity.

Conclusion validity:

Conclusion validity involves comparing the results of the study and the conclusion, for example if in a study we want to establish the relationship that exits between variable A and B and that A causes B then if we conclude that A causes B then the results are valid.

Internal validity:

In this form of validity we establish whether the relationship between variables is causal, example if we establish that A and B are related in a study we test whether A causes B and this is achieved through tests.

Construction validity:

This validation method tend to establish whether the relationship being established is credible, this is done through determining whether the measures that were intended to be measured were measured, this can only be achieved if the study construction was well prepared for the

Social Science Research

collection of observations.

External validity:

This tends to establish that a study is valid if same results can be achieved using another sample or if the study was undertaken at different area or different time period, this is what is referred to as external validity. Therefore All the above validity methods will help establish whether a research proposition or inference is valid or not valid.

validity and reliability:

in social research we can have both valid and reliable results, we can also have reliable but invalid results, we can also have valid and non reliable results and finally we can have invalid and unreliable results. We take an example retrieved from which describes the various cases a social research can take, in this example we assume that we are trying to get certain information from research, we assume that when we get accurate results then we are hitting a target at the center and the more we get inaccurate results the more we are missing the target, this diagrams shows levels of reliability and validity:


Social Science Research




we can discuss the diagram as follows:

Diagram A shows reliability but results are not valid, in such a research study we aim at measuring a certain variable but we end up having results that are reliable but not valid.

Social Science Research

Diagram B shows results that are valid but not reliable, the results are valid because they are distributed but they are unreliable because results are not consistent.

Diagram C shows a situation in which we do not have valid and still the results are unreliable, finally in the last diagram D shows reliable and invalid results.


Social research methods (2008) validity and reliability in research, retrieved on 13th May, available at

Psychological construct(2002) shyness measure research, retrieved on 13th May, available at h ttp://