Skip to main content

Posts

Showing posts from September, 2017

AGE SCORE PROBLEMS

What are age scores? Age scores, also called age-equivalent scores , are supposed to help people understand how a person’s test score compares to other people of the same age. They are often provided to teachers and parents to show how children scored on achievement tests compared to their age peers. A common age-equivalent abbreviation is AE .  Age scores are reported with a hyphen. The first number refers to age in years and the second number refers to the age in months. A score of 8-4 is supposed to mean a test performance typical of children age 8-years and 4-months. The scores appear convenient and make a kind of common sense. An age score of 7-6 is supposed to mean that a child earned a test score similar to children age 7-years and 6-months. But there are problems with the scores. What tests report age scores? Age scores are commonly reported with results of achievement tests. They are sometimes reported with results of intelligence tests. Old intel

Measuring Spiritual Outcomes in Psychotherapy

The Theistic Spirituality Outcome Scale (TSOS) has potential as a useful outcome measure. Recently, a group of us completed a study of clients who saw Christian counselors. We assessed their current well-being using two measures: The Schwartz Outcome Scale (SOS) and the Theistic Outcome Scale (TSOS). (See references below.) The TSOS was designed by Richards (2005) as a measure of well-being for people associated with a theistic religion like Christianity, Judaism or Islam. We used the 17-item version, which uses a 5-point response format from  1 = never to 5 = almost always to rate each item (e.g., “I felt spiritually alive.”). Reliability We only calculated coefficient alpha, which was strong at .95. Validity The TSOS was significantly correlated with ratings of satisfaction with Christian counseling (.65) and likelihood of returning to Christian counseling (.62). It was significantly correlated with the SOS measure of general well-being (.84). Other significa

Reporting Mean or Median

Who would think that a simple statistic like a mean or a median would make a difference? In large samples involving thousands of people, and when data are normally distributed (close to the shape of a bell curve), the mean and median will be nearly the same. In fact, in a theoretical distribution called the normal curve , the mean , median , and   mode are in the middle. But, many samples are not normal distributions . Instead, the often contain extreme scores called outliers or a lot of scores bunched up at high or low levels ( skewed ). Sadly, even people that understand statistics, continue to report the mean as if they are not thinking about their samples. Suppose you work for a company where the top person earns $300,000 but most folks earn $30,000 to $60,000. Well that $300,000 is gonna skew results and the mean will look much higher than the median. I ran some fictitious data on a sample of 10 people. Nine earn between $30 and $60K and one earns $300K. The Mean = $6