Percentile Ranks

¶ … person's ability/skill/standing relative to other people in order to design an intervention (Cohen & Swerdlik, 2013; Runyon, Coleman, & Pittenger, 2000). Summary statistics (descriptive statistics) allow for the assessor or researcher to summarize the raw data and scores from people being tested. The percentile rank or percentile score represents a person standing relative to other individuals on a particular test (Cohen & Swerdlik, 2013). There are some drawbacks to percentile scores.

Obviously if the percentile rank is obtained without consideration for the data then the percentile rank may not be a meaningful or accurate summary statistic. For example, if one uses a standard distribution such as a Z. distribution to determine the percentile rank of a particular score and the distribution of raw scores is not normal then the percentile rank will not be an accurate reflection of the person standing on that particular measure (Huck, 2012).

Perhaps, one of the biggest mistakes made in assessment and testing is to treat highly skewed distributions of scores as if they were normal distributions (Cohen & Swerdlik, 2013). This actually happens often on physiological measures such as measures of reaction time, which is typically a positively skewed distribution where the majority of people being tested score at the lower end of the distribution and fewer people scoring at the higher-end (Runyon et al., 2000). Transforming a highly skewed distribution into a Z.

distribution results in an inaccurate representation of the scores (Cohen & Swerdlik, 2013). However, just as a raw score on a test is often meaningless without some comparison, a percentile rank is also meaningless in the abstract (Huck, 2012). For example, if someone claims to have scored at the 99th percentile on a test of aptitude for mathematics this particular statement is relatively meaningless. Suppose that the reference group on this particular test is composed of people who only completed the fourth grade. Knowing the reference group is extremely important.

If someone claims to be at the 99th percentile on a test of mathematics aptitude that might sound impressive until you learn that the reference group for that test is fourth-graders. Thus, simply stating that one has scored at a particular percentile rank is meaningless without knowing the reference group. The biggest drawback in interpreting percentile scores is interpreting them in light of an inappropriate reference group or no reference group at all (Cohen & Swerdlik, 2013; Huck, 2012).

Depending on the type of test demographic variables such as age, education, gender, etc. can make a big difference in where one's standing relative to others on the same measure lies. It is important to make sure that the appropriate reference group is being used when calculating percentile scores (Cohen & Swerdlik, 2013). This is why many the developers of standardized tests that are employed in education, psychology, and industry publish separate norms for different reference groups (Cohen & Swerdlik, 2013).

A high percentile score compared to one reference group may actually represent a low percentile rank score on another reference group. In the above example knowing that someone scored at the 99th percentile on a test of mathematics aptitude where the reference group is comprised of fourth-graders is not nearly as impressive or meaningful as scoring in the.