Validity
Validity is the degree to which an instrument measures what it is supposed to measure. Validity is also the extent of concordance between a measure and an underlying theoretical variable. Four types of validity, face, content, criterion, and construct, are presented in Table 4.1.
All evaluation must establish the validity (accuracy) and reliability (reproducibility) of data collected in measuring the eight categories of data noted in the introduction.
Face Validity
Face validity, a dimension of content validity, describes the extent to which an instrument appears to measure what it is supposed to measure. Thus, the question, “How many minutes of exercise did you do today?” appears to measure one important component of aerobic activity for a particular day. This question may or may not produce good data for adults. This question would not be an accurate estimate of physical activity duration if young children were asked this question.
If an evaluation needs to assess aerobic activity that has cardiovascular benefit, a person must engage in aerobic activity for > 20 minutes at a time (without stopping), for > 3+ times every week. The activity must reach a certain intensity level, a heartbeat > 60% of a maximum rate, calculated from commonly available tables for age and gender groups, to promote
|
type |
defi nition |
|
Face |
The extent to which the instrument appears to measure what it is supposed to measure. |
|
Content |
The extent to which an instrument samples items from the full range of content. |
|
Criterion |
The extent to which a new instrument correlates with another more accurate (and usually more expensive) instrument (the criterion). |
|
Concurrent |
The extent to which the scores from two instruments or sub-scales are correlated. |
|
Predictive |
The extent to which an instrument administered during one time period can predict changes in a prospectively assessed criterion measure or rate. |
|
Construct |
The extent to which the measure of concern correlates with other measures in predicted ways, but no true criterion exists. |
|
Convergent |
The measure correlates with items with which it is predicted to correlate. |
|
Discriminant |
The measure does not correlate with items with which it is expected not to correlate. |
CVD fitness. Thus, it may be difficult to interpret an intensity response from the above.
Suppose the answer a person gives is “> 30 minutes.” Was the activity sufficiently intense to merit the label “aerobic”? Did the person cover 4 miles (excellent CUD benefit) or 2 miles (modest CUD benefit) in 30 minutes? Were the 30 minutes in one block or two 15-minute segments? Each variation raises issues about whether the question elicits accurate data. You will need to create a self-report measure for adults to assess distance traveled, intensity (e.g., heart rate), and continuous duration of activity (segments in minutes). A measure, therefore, must be carefully written to assess impact. Face validity is a small first step toward overall validity of the instrument. As there is no quantitative measurement to face validity, experts are consulted.
