The questions that make up the WHOQOL-100 arose from a process designed to capture both the culture-specific interpretation of quality of life facets as well as language idiom. There was therefore, of necessity, some diversity in the nature and structure of the questions. Consequently, there was a trade-off between a minimum number of standardised question-response scale formats whilst still allowing an enquiry into difficult aspects of quality of life, and maintaining the unique face validity of the questions in the WHOQOL-100 in different cultures. To accommodate this there are four five-point response scales concerned with the intensity, capacity, frequency and evaluation of states or behaviours.
The Intensity response scale refers to the degree or extent to which a person experiences a state or situation e.g. the intensity of pain. Questions may also refer to the vigour or strength of a behaviour. The assumption is that the experience of a more intense state is associated with corresponding changes in quality of life. Example questions include: “Do you worry about any pain or discomfort?” and “Do you have any difficulties with sleeping?”. One response scale is used to assess intensity. In English, the anchors on the scale are “Not at all” and “Extremely” or “An extreme amount”.
The Capacity response scale refers to a capacity for a feeling, state or behaviour. The assumption is that a more complete capacity is associated with corresponding changes in quality of life. Example questions include: “Do you have enough energy for everyday life?” and “To what extent are you able to carry out your daily activities?”. In English, the anchor points are “Not at all” and “Completely”.
The Frequency response scale pairings refer to the number, frequency, commonness, or rate of a state or behaviour. The time frame is crucial to these questions, such that the frequency refers to its frequency in the specified time period. The assumption is that a greater number of occurrences of the state or behaviour is associated with corresponding changes in quality of life. Example questions are: “How often do you have negative feelings, such as blue mood, despair, anxiety, depression?” and “How often do you suffer (physical) pain?” In English, the anchor points are “Never” and “Always”.
The Evaluation response scale refers to the appraisal of a state, capacity or behaviour. The assumption is that a more positive evaluation is associated with a corresponding increase in the respondent’s quality of life. Example questions are: “How satisfied are you with your capacity for work?” and “How satisfied are you with your personal relationships?” Several evaluation scales are employed. In English, the anchor points are “Very unhappy” - “Very happy”; “Very dissatisfied” - “Very satisfied”; and “Very poor” - “Very good”. This response scale differs from the intensity, frequency and capacity scales in that it has as a neutral midpoint and the anchor points are not extreme points, to maximise full usage of the scale. In several languages (e.g. Croatian and Dutch) the distinction between the two question stems “How satisfied...?” and “How happy...?” does not translate and all of these questions and response scales therefore become “How satisfied...?”
Response scales have been derived for each of the WHOQOL-100’s language versions according to a standardised methodology. Ensuring equivalence in response scales required a methodology that goes beyond translation of standardised English language scale descriptors. Although endpoints such as “Never” and “Always” are universal, shades of meaning between endpoints (e.g. “sometimes”) are more ambiguous, difficult to translate, and subject to cultural variation in their interpretation. To ensure equivalence across WHOQOL field centres, a methodology was used which specified the anchor points for each of the four types of 5-point response scales (Evaluation, Intensity, Capacity and Frequency), and a scale metric which intermediate descriptors should fit. That is, descriptors for each of the response scales were derived to find words/terms falling at 25%, 50% and 75% points between the two anchors.
This methodology ensured first that response scales were not simply translated from a source language with all the problems associated with this process. Second, it secured a high degree of scalar equivalence between languages, which was supported by subsequent bilingual review. Third, it ensured equidistance between descriptors on the scales. The method whereby response scales were derived is described more fully elsewhere (Szabo, S., Orley, J. and Saxena, S. On behalf of the WHOQOL Group, 1997).