The conclusions drawn, whether from survey analysis, experimental design, or any other research approaches, are useful only if they are valid. One has to obtain answers to “How true are the findings?, How well do they represent or support the concept you are studying?.” This is when validity comes into the picture.
Validity determines whether the research measures what it is intended to and approximates the truthfulness of the findings.
Validity, segregated into internal, external, content, face validity, etc. is impacted by several factors.
Some of the factors that impact internal validity are subject variability, size of subject population, attrition, time given for the data collection, sensitivity of instrument, etc. On the other hand, external validity gets affected by aspects such as population characteristics, interaction of subject selection, descriptive explicitness of the independent variable, data collection methodology, effect of time, research environment, etc.
Among various types, the most commonly used validity include:
- Face validity
Face validity refers to the degree to which a test seems to measure what it claims to measure. Face validity, considered as a minimum and basic index of content validity, is determined after the construction of the test. It is the simplest and least precise approach of determining the validity and relies entirely on the expertise and familiarity of the scholar. Face validity ensures that the measure appears to be evaluating the intended construct under study. However, it is generally utilised to describe the appearance of validity without performing empirical tests. For instance, estimating the speed of a bike based on its outward appearance.
- Content validity
This type of validity is the extent to which the questions on the instrument and the scores from the questions provide possible questions that can be asked about the content. Content validity makes sure that the questionnaire consists of necessary set of items to tap the concept. Content validity depends on the judgement of experts in the subject domain and is interested in evaluating present performance rather than predicting future performance. Here the unclear and obscure questions can be modified, and ineffective or non functioning questions can be removed. For e.g., if we want to test knowledge on Indian Geography, you cannot have questions limited to the geography of Delhi, the capital of India. Upon determining the truthfulness, in need for help in penning down the validity section in your dissertation? Consult UK dissertation writers.
- Criterion-related validity
This kind of validity is used to predict current or future performance. Criterion-related validity correlates test findings with another criterion of interest and deals with the relationship between scale scores. It examines how the scale alters individuals on a criterion it is expected to predict. That is, when we the future performance is expected based on the scores that are currently obtained by the measure, it correlate the scores acquired with the performance. For instance, a hands-on driving test is considered to be an accurate test of driving skills. The test may be repeated by the paper-pen test to compare the validity of the test.
Validity is a mandate for research in all domains. Although the errors in validity cannot be eliminated completely, achieving 95% of truthfulness of results can make your research a potential one. If you need some help in including the same in methodology section, take aid from the best UK dissertation writers.