Examine The Need For Validity And Reliability Business Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Psychological research shows that references and interviews are inaccurate selection methods. "Subjective measures, like interview assessments, often aren't so consistent. At their worst they may be so inconsistent that they convey no information at all. Mark Cook (2001) page 8. "Accuracy divides into two issues "RELIABILITY AND VALIDITY" (ibid).

Validity has a lot of meaning which differ from many authors around the globe. It has no fixed perception. We will look at Validity from different perspective. "Validity" ask questions like 'is the means of the assessment actually accurate or not, and also whether the assessment is actually measuring what it ought to measure? "The factors measured should be relevant to the effective job performance" (Wendy Banfied 2008).

Does the test measure what it claims to measure? Does the information obtained from it enable you to make relevant judgement about the person? (Bartram, etal January 1990) page 17. Colman (2001) described Validity as the extent to which specified inferences from the test score are justified or meaningful. It is measured by correlation coefficient between scores on test and criterion Validity (2001: pg 781).

A renowned definition is that an "Account is Valid or True if it represents accurately those features of the phenomena, that it is intended to describe, explain or theories" (Hamersley in 1987) (1). According to Mark cook (2001) pg 210 " A Valid test is one that works, measures what it claims to measure that predicts something useful." 2. Kerlinger page 444-445 in 1964) in his own stated "Are we measuring what we think we are?". Dunnette (1966) added that Validity refers "Broadly to the process of learning more about the meaning or total network of interpretation that may be attached to an individual different measures", it involves learning more about the meaning of a test; A valid test is backed by research and development. "Two important means of establishing the validity of a selection instrument are the statistical and the content methods" Gregorio B (1999)


As we say in our every day discussion, that something is reliable, a friend is reliable, news paper television is reliable, that means they are dependable or trustworthy! But for the purpose of this report, it will go beyond being dependent or trustworthy, so here it will include being consistent and repeated. That is to say that any particular test which would give you an answer, the same answer over and over, for instance if a student score 60% in a job aptitude test and the test was repeated the same student score the same 60%, it means that they was consistency in the marks obtained. "What relevance can you place on the score that somebody obtains for psychological tests?", this precision of measurement is what is referred to as Reliability. (D. Bartram, P. A. Lindley and J. Foster).pp 17. 1990. However, there are two parts in this score, one is "True Score" and the other is the "Error" True Score is the level the candidate actually got, while the 'Error' is the Error which occurred during marking. "A test is 'stable' if people tend to get the same score each time they take the test." These errors can occur as a result of the condition which the exam was taken. For example if a candidate was unwell during the period of the test, then the candidate actual performance will not be reflected in the test. "The items which make up a test scale are all designed to measure the same attribute, we can say that a test provides a constituent measure of an attribute if the responses given to each item or question are related"(ibid). Wendy Banfield (2007) states "is the selection method consistent in the results it yields? …. Reliable across different selectors? …. Reliable if re-taken by the same applicant?.

Let's look at different definitions of Reliability cited by different authors; Reliability is "Ability to measure consistently". (Black and Champion) page 234-236 in 1976. Another definition is that Reliability is "Accuracy or Precision of a measuring instrument", (Kerlinger 1964, page 430-444). Johnston and Pennypaker cite that Reliability is "Capacity to yield the same measurement…. Stability" (1980 page 190-191). Colman (2001) states that "Reliability is the internal consistency and stability with which a measuring instrument performs its function" this has to do with the measuring instrument producing the same result when used on the same candidate on different occasion.


VALIDITY AND RALIABILITY are actually interlinked to each other, or lets say there are related with each other. For any selection process to be Valid it must be Reliable too, meaning that the process must measure what its meant to measure, and do that repeatedly, since Reliability is measured in terms of selection process performance and Job performance, if the process is unreliable then the process will definitely not be Valid. According to Chris Barker etal (2002) "Reliability is a necessary but not sufficient condition for Validity." "To be Valid, a measure must first be Reliable; otherwise it would consist mainly of error." On the other hand, a measure can be highly Reliable but still not Valid. (Ibid). To me I feel that Reliability is an essential part of Validity, because Validity does not measure only the positive outcome of a selection process, but also how Reliable the selection process is?.


Selection is the process used to find the best candidate for the organization. It considers a lot of factors like the candidate's skills, knowledge and ability.

In selection, we seek to predict the latter job behaviour from the results of measures administered when candidates apply for the job. According to (Hackett) Selection is concerned more with

"Predicting which candidates will make the most appropriate contribution to the organization now and in the future".

To ensure that the right candidates are selected for the required job specification, it is essential that right candidates are selected. If the wrong candidates are selected, it will lead to sub-standard performance, which will adversely affect the organization. Selection involves a number of methods such as interviews, biography, data, personality inventories and assessment centre evaluations. These methods are all dependent on standards such as Validity and Reliability. Organisations have to carry out studies to make sure that the various methods used are "Valid" or not, A selection process may have an unfair impact on a particular group say the minority, fro example if an organisation advertises for a particular Job for higher qualification say MBA or a certain number of years or certain number of experience in a certain job, the selection requirement will obviously have an adverse effect on the minority group since, they will be less number, and will not be as many as the majority group. Another major procedure to be considered by organisations is "Job Analysis." According to SIOP (2002), "It is the foundation for the development of any selection procedure." A job analysis reveals the critical tasks for a job, and the relevant knowledge, skills and abilities, (KSA's). Asking a candidate question that does not relate to a job he or she is applying for will definitely create error in the whole process, so to improve the validity of a selection process, they should be a designed Job-related questions or tests, given to all applicants of that particular job, this will help to reduce error and bias.

Another important factor is "Job Description", this is a detailed statement of what the job entails, identifying the task involved and the skills, and responsibilities involved, this will allow the candidate to know actually what the description of the job is, job title, the department, purpose of job and objectives, physical and economic conditions of the job, this will serve as a determinant of the qualities a candidate must posses in other to be successful, this will also allow give the candidates a clear understanding of the job.

Another selection tools organisations use is "Interview" An interview is a "formal face to face meeting, especially, one arranged for the assessment of the qualification of an applicant, as for employment" www.dia-sho.com/pgsa2/p9sa-glossary.html. Although it is the structure of the interview that determines the reliability of the process. "Research has suggested that without proper care, interviews can be unreliable, low in Validity and biased. Graves L. and Karren, R. (1996). Content interviews are more reliable, content focused and measurable.


Other practical imperatives affecting selection include "Work Experience" the type of industry and trade the applicant is coming from should be considered in relation to the Industry and trade he is applying. Also the Salary package you are offering for the vacant job and what the candidate is asking for, the applicant's education and the BFOQ. (Bona Fide occupation qualification), this is a justifiable qualification which allows the employee to select candidates from a particular group of people, say, Religion, Sex, Community or Age, in some countries like Northern part of Nigeria and Pakistan.

More so, a candidate who is coming from a particular industry, would have been used to the culture of that particular industry, so it will take a while to adjust to other industry not minding his expertise in his field, he still need some time to adopt to the other industries norms, values and atmosphere.

However, if the salary the organisation is ready to pay for an advertised job is 5,000 pounds, and during the interview the interviewer found out that the applicants previous salary was 3,000 pounds, they might not acknowledge the candidate, because they would expect that his previous salary should be within the range of 4,000 - 4,500 pounds, Because this may affect inner justice and balance regarding pays. The organisations internal and external balance should be maintained or else one day, some one would come to say that, some one with the same amount of experience is getting more pay.

Another factor to be considered is "Related Experience" If an organisation advertise for a particular job in a particular Department with a 5 years experience in that department only, and some one with 7years experience in related field turns up, Human Resource Manager will see it as taking a risk, because what the organisation want is a candidate who has been working in that particular department and no where else.



One major factor which can affect selection processes is the contract agreement with Labour Union, this agreements often affect the way selection is done, and one typical example according to Bhaskar Chatterjee is that "Unions frequently resent 'lateral' entry They tend to protect the interest of their members, and would want vacant positions at the top to be occupied by their members in the lower hierarchy. Labour Union will always try to persuade the management to keep to their agreement of selecting people from lower hierarchy to occupy certain position when there are vacant, thus affecting the selection process. "There is always nearly always a tussle between the Union and the management as to the number of and / or percentage of staff that may be recruited in a particular year…………….. Such tussles result in agreements and understandings which in turn circumscribe the scope of recruitment" Bhaskar Chatterjee third edition (2007) page 106. In some cases where there are Government regulations concerning selection where a certain Group, Tribe, or Gender in some countries, it will definitely have a direct impact on the selection process.


There are factors which are external which can hinther selection process, first are that there are higher chances of making mistake by the new staff, since he is new to the organisation. In organisations, more so, he may be too slow at the start. There are policies there are bound to affect the way selections are being carried out in such organisations. For instance where an organisation encourages only promotion from within their organisation. This type of situation causes problem in an organisation, because workers work only to be promoted while there is no one bringing in new expertise, and there is always favouritisms. But the fundamental fact is that whatever the declared / undeclared policy or practices - they have significant role in determining who is recruited.


Selection involves a number of methods such as interviews, biography, data, personality inventories and assessment center evaluations. There are several processes that can be adopted to ensure companies are selecting the right person such as varying forms of interviews, psychometric testing and assessments among others. However, these forms of selection tools are dependent on their reliability and validity as to the extent to which they can be of benefit for companies to use All these selection procedures have different methods of measurement to assess a variety of characteristics. Before any of these processes can be used certain standards need to be met to ensure processes are accurate. Five critical standards with which all employee selection methods should conform are reliability, validity, generalisability, utility and legality (De Cieri & Kramer, 2003).

Panel interviews are another way of decreasing bias. There are several forms of structured interviews being used in selection methods. Some of the most common are Situational Interviews (questions centring on what action candidates would take in various job-related situations), and Behavioural Interviews (candidates are asked what actions they have taken in prior job situations that are similar to situations they may encounter on the job). Behavioural interviews are quite high in reliability because past behaviour is the best predictor of future job performance (Russell, 1999).