Test Development

Discuss the process you would follow in developing a psychological assessment measure. Discuss the steps that you would take in this process, including how you would choose items for your test, how you would evaluate the reliability and validity of your test, and the issue of establishing norms.

Developing a test.

Developing a test begins with careful planning. The first and most important step would be to establish the aim of the test, for example, to test artistic skills in small children as an expression of creativity. It would be during this stage that a researcher would establish what the measure would be used for, such as establishing levels of creativity within children as a means of assessing a different form of intelligence for children who battle within a mainstream classroom. The population the test would assess would be considered during the planning phase. In this example the test would be used to assess primary school children. (Foxcroft & Roodt: 2001).

A planner would also consider which decisions would be made based on test results, and how to establish norms for the test, so that ,for example, the creative ability of any one child is compared to a group of children of similar age. This means that a child's assessment is made more meaningful. By comparing it to the abilities of children of similar age, a child's abilities are shown to be lesser or greater than average, showing whether a child would excel or flounder within a creative schooling environment. A test could also be ipsative, which means that it would be based on an intra-individual comparison. The creative levels of a child could, for example, be assessed before and after remedial games and play, and his results could be compared to previous ones as a means of assessing progress. (Foxcroft & Roodt: 2001).

After planning the goal of the test, it would be important to define the concepts, or constructs, such as creativity, which the test would aim to assess. Creativity is a hidden concept because ultimately it can only be seen by looking at creative behaviour. We cannot look at a person and assess his creative ability if that person does not show some form of behaviour which would indicate either a creative ability of lack thereof. When we assess creativity as a hidden construct, we would need to do so by explaining creativity through observable behaviours. This is called operationally defining the construct. When planning to test creative abilities within children, we could measure tasks, for example, which reflect musical or artistic ability by letting children sing or play an instrument, or assessing drawings and use of colour, or how a child can create using play dough or wooden blocks. These tasks, which can be observed, would operationally define the construct of creativity. (Foxcroft & Roodt: 2001).

During the final planning phase, it would be considered as to how a test would be presented. In assessing small children, a test may include bright and attractive products which would catch a child's attention, and activities would be planned around the short attention span of young children. When testing with questions, certain guidelines would be considered. Questions could be open ended, which could mean asking a child, for example, whether he enjoyed painting or pretending to be a cartoon character. Such questions could give a lot of information but also be a little problematic in that the answers may take up a lot of time. A test may also include questions with prepared answers, such as true or false, or multiple choice answers. A child could answer whether it would be true or false that he enjoys painting or drawing. He could also answer a multiple choice question, which would give options, such as always, sometimes, a little bit or hardly ever . Children could complete a task such as finishing a sentence. Drawing is..... Older children could complete a story and children could play with toys or apparatus which would indicate creativity. The time the test would take would need to be taken into consideration, as young children become restless and tired easily. Scoring the test would also need to be easy to understand for those who would administer the test. (Foxcroft & Roodt: 2001).

Assembling the test

Writing up test items, either as instructions or as questions, would be the next step in test development. Foxcroft & Roodt (2001) point out guidelines to ensure that these are as clear and uncomplicated as possible. Such instructions include avoiding negative sentences, such as 'I never draw' or double negatives, such as 'it is not often that i do not draw' which is confusing and difficult to understand. Using only one topic per sentence is also important, as it helps to make answers more easy to give. A true or false question such as 'I like to draw' is more easy to answer than 'I like to draw, paint, sing and act'. Questions which are vague are not helpful to test construction. Asking, for example, whether a child likes artistic activities is vague, because artistic activities could be interpreted as only those things which are done in the art classroom with the boring teacher, and a child may give an incorrect answer.

In order to ensure that questions are read properly, instead of all boxes on, say, the right are ticked, Foxcroft & Roodt (2001: p74-75) suggest varying the positioning of answer boxes so that test takers have to read all questions. It is also suggested that true and false answers are the same length, and that there are an equal number of true and false answers. Children should be able to use apparatus easily and should find the apparatus attractive to use.

Reviewing the test

Now that the test has been thought out and designed, the task is to prepare the test for experimentation on a sample of people who represent the population of eventual test takers, such as children. It is also time to draw up a test booklet, and to determine whether the answer sheet should be a part of the booklet, or separate. Answer sheets should be easy to use so that the person who administers the test does not make mistakes. (Foxcroft & Roodt: 2001).

Length of the test is important. The more items a test includes with regards to a construct such as creativity, the more insight a tester will gain into the creative nature of the child. However, if a test is too long, it can create fatigue and give less than accurate results. It is important therefore to balance the items for assessment with a helpful time period for the test. (Foxcroft & Roodt: 2001).

Administrators