Skip to Content

What do T scores mean?

T scores are a commonly used statistic that allow you to compare an individual’s performance on a test to the performance of a relevant population or norm group. T scores have a mean of 50 and a standard deviation of 10. This makes them easy to interpret – a T score of 50 is average, while higher and lower T scores indicate above or below average performance compared to the norm group.

What is a T score?

A T score is a standardized score with a mean of 50 and a standard deviation of 10. The ‘T’ stands for ten, referring to the standard deviation of 10. T scores are a type of z-score, which are standardized scores with a mean of 0 and standard deviation of 1. To convert a z-score to a T score, you multiply the z-score by 10 and add 50. This shifts the mean to 50 and the standard deviation to 10, making T scores easier to interpret.

The formula to calculate a T score is:

T score = (z-score x 10) + 50

Where:

  • z-score is the individual’s score expressed as standard deviations from the mean
  • 10 is the standard deviation of T scores
  • 50 is the mean of T scores

Some key properties of T scores:

  • The mean is always 50
  • The standard deviation is always 10
  • T scores follow a normal distribution, shaped like a bell curve
  • About 68% of T scores fall between 40 and 60 (within 1 standard deviation of the mean)
  • About 95% of T scores fall between 30 and 70 (within 2 standard deviations)
  • T scores below 30 or above 70 are considered statistically unusual

When are T scores used?

T scores are commonly used when reporting results from educational or psychological assessments. Some examples include:

  • Standardized academic achievement tests like the SAT, ACT, GRE
  • Intelligence and cognitive ability tests like IQ tests, the WISC, WAIS
  • Personality tests like the MMPI
  • Neuropsychological tests
  • Career/vocational interest assessments like the Strong Interest Inventory

T scores allow easy comparison to the norm group. For example, on an IQ test with a mean of 100 and SD of 15, an IQ of 130 would be 2 SD above average. Converted to a T score, it would be a 70, easily indicating it is 2 SD above the mean.

Interpreting T scores

When looking at a T score, the main thing to consider is how it compares to the mean of 50:

  • T score above 50 – Performance is above average compared to the norm group
  • T score of 50 – Performance is average
  • T score below 50 – Performance is below average

More specifically:

T Score Range Performance Level
70 and above Very high
60 – 69 High
45 – 59 Average
35 – 44 Low
34 and below Very low

T scores between 40-60 are considered within the average range. Scores below 30 or above 70 are unusual and may need additional assessment or testing.

Advantages of T scores

There are several advantages to using T scores rather than raw scores:

  • Standardized – Can compare scores from different tests/scales
  • Mean of 50 – Simpler to interpret than scores with varying means
  • Shows norm comparison – Demonstrates performance relative to peer group
  • Normal distribution – Conforms to bell curve, consistent percentiles
  • Equal units – Each 10 point difference indicates 1 standard deviation

The standardization and fixed mean and standard deviation allow for easy comparisons across tests and norm groups. This makes T scores popular in educational and psychological testing.

Disadvantages and limitations

Some drawbacks and limitations to consider with T scores:

  • May obscure very high or low scores due to standardization
  • Assumes normal distribution, though data may not fit
  • Difficult to interpret for skewed distributions
  • Cutoffs can seem arbitrary (e.g. 40 vs 39)
  • Does not indicate absolute level of performance

T scores standardize scores so they may downplay more extreme high or low scores. The fixed mean and standard deviation also assume the data closely follows a normal bell curve, which is not always the case. Additionally, T scores alone do not reveal actual skill level or abilities.

Examples of T scores

Here are some examples to demonstrate calculating and interpreting T scores:

IQ test

On an IQ test with a mean of 100 and standard deviation of 15, a child receives a score of 130. What is this as a T score?

First convert to a z-score: (130 – 100) / 15 = 2
Then convert the z-score to a T score: (2 x 10) + 50 = 70

A T score of 70 indicates the child’s performance is very high compared to peers.

Academic achievement test

On a standardized math test with a mean of 200 and SD of 40, a student receives a raw score of 160. What is the T score?

(160 – 200) / 40 = -1 z-score
(-1 x 10) + 50 = 40 T score

A T score of 40 is slightly below average performance compared to other students who took the exam.

Personality assessment

On a personality assessment measuring extraversion with a mean of 15 and SD of 3, a person receives a raw score of 10. The T score is:

(10 – 15) / 3 = -1.67 z-score
(-1.67 x 10) + 50 = 35 T score

A T score of 35 indicates the person scored low on trait extraversion compared to the norm group.

Using T scores in research

In scientific research and statistics, T scores are commonly used when comparing groups or measuring change over time. Some examples include:

  • Comparing experimental and control groups
  • Examining pre-post change in assessment scores
  • Evaluating differences between treatment interventions
  • Looking at variations across demographic factors

Converting raw scores to T scores allows for standardized comparisons on equal units of measurement. Researchers can then use parametric statistical tests like T-tests and ANOVA to analyze the T score data.

Research example

In a drug trial, researchers administer an experimental drug and placebo to randomly assigned groups. They measure depression using a scale with a mean of 100 and SD of 15. The results for each group are:

Group Mean Raw Score Mean T Score
Drug 75 35
Placebo 95 45

Converting to T scores allows the researchers to easily compare the standardized means and show the drug group scored 1 SD lower on depression than the placebo group.

Comparing T scores to other standardized scores

While T scores are a commonly used standardized score, there are a few other standardized scores that are used in similar contexts:

  • Z-scores – Have a mean of 0 and standard deviation of 1
  • Stanines – Scores from 1 to 9 with a mean of 5
  • Stens – Scores from 1 to 10 with a mean of 5.5
  • Percentiles – Rankings from 1st to 99th percentile

Compared to these other formats, T scores have the advantage of a more intuitive and easier to interpret scale centered on 50 as average. The equal units of standard deviation and defined cutoffs also help with norm comparisons. However, other standardized scores like percentiles may sometimes be more understandable to those without a statistics background.

Conclusion

In summary, T scores are a useful standardized score that allow individual results to be compared to a relevant reference group. A T score of 50 represents the mean, while higher and lower scores indicate above or below average performance compared to the norm. T scores are widely used in educational and psychological testing due to their advantages of standardization, normal distribution, and ease of interpretation. However, they also have limitations such as obscuring more extreme scores. Understanding how to properly interpret T scores provides valuable context for evaluating test results and research data.