Self-Report Techniques - Validity, Reliability and Ethical Issues

Validity External ValidityIn interviews and questionnaires this concerns the extent to which the findings can be generalised to other situations and other people. A major factor will be the representativeness of the sample used to collect data.E.g. If a questionnaire collected data only from shoppers on a weekday morning in London it's not reasonable to generalise this to all people in the UK.Internal Validity In self-report techniques it is related to the issue of whether the questionnaire or interview really measures what it intended to measure. The most common ways to assess this is:1. Face Validity. Does the test look as if it is measuring what the researcher intended to measure. In order words, are they relevant and related to the topic?2. Concurrent Validity. This can be established by comparing the current questionnaires or test with a previously established test on the same topic. Participants take both tests and then the 2 test scores are compared.

ReliabilityInternal reliability This is a measure of the extent to which something is consistent within itself. E.g. All the questions on a IQ test should be measuring the same thing.However internal consistency isn't always important, say for example, a questionnaire about day-care experiences might look at many different aspects of day care and its effects.External reliability This is a measure of consistency over several different occasions. E.g. If an interviewer conducted an interview, and then conducted the same interview with the same interviewee a week later, the outcome should be the same - otherwise the interview is not reliable. Reliability also concerns wheather two interviewers produce the same outcomes, this is called inter-interviewer reliability. Assessing reliability:Internal reliability - Split-half methodA single group of participants all take a test once. Their answers are then divided in half. This could be done using odd or even numbers or just half the questions.The individual's scores on both halves should be very similar. The two scores can be compared by calculating a correlation coefficientExternal reliability - Test re-test methodA group of participants are given a test or questionnaire or interview once and then again sometime later. The answers can be compared and should be the same. If the tests produce scores, these can be compared by calculating a correlation coefficient. Improving Reliability-Removing those items which are most inconsistent. The only way to do this is by trial and error - remove one item and see if the spilt-half correlation coefficient improves. If it does, then the removed item should be permanently left out.

Ethical Issues- Deception about how true research aims may sometimes be necessary in order to collect truthful data.-Psychological harm , respondents may feel distressed by certain questions or having to think about certain sensitive topics.-Privacy questions may be related to sensitive and personal issues, invading an individuals's privacy. - Confidentially must be respected; names and personal details shouldn't be revealed without permission. No personal data may be stored. 

!