
A VERY fascinating aspect of research is questionnaire validation. And by its nature it can be madugo, a Tagalog term that the English literal translation “bloody” cannot quite capture.
Your questionnaire must be acceptably sensitive (detects what it is designed to detect); positively predictive (predicts what it is designed to predict); and negatively predictive (does not predict what it is not designed to predict).
You also have the privilege of understanding better the behavior you want to study later on. A case in point is a study by Rachel Vilberg, Ingrid Munck and Nancy Petry that the American Journal of Addiction published in May last year. The study aimed to enhance screening capability of the Nods-Clip questionnaire, itself an enhancement of the National Opinion Research Center (Norc) DSM-IV Screen for Gambling Problems. (Nods refers to Norc Diagnostic Screen for gambling disorders.)
The Nods has a sensitivity of 98 percent for lifetime pathological gamblers and 90 percent of lifetime problem gamblers. Using three criteria of loss of control, lying and preoccupation (Clip), the Nods-Clip detects 99 percent of Nods pathological gamblers and 94 percent of Nods problem gamblers. But it cannot discriminate at-risk and low-risk gamblers.
So Nods-Perc was conducted to correct this. It used the criteria of preoccupation, escape, risked relationships and chasing (or Perc). And it did. It is 99.7 percent sensitive (very high); 88.5 percent positively predictive (better than Nods-Clip); and 96.3 percent negatively predictive (far better).
I am bringing this up in Breakthroughs because the findings gave an exceptional insight into the gambling behavior. If you have gambling problems, it will be an opportune time for self-diagnosis.
If you answer “No” to all the four criteria below, consider yourself... [READ MORE]
This article appears in SunStar Cebu newspaper on 12 December 2012.
Comments
Post a Comment