Testing Construction_ Module 61-3

The world-wide observed increase in standardized test scores throughout many years is called the ____ ____.

When you ask a test group to repeat the measure multiple times to ensure consistent results.

Ally's watch will ALWAYS be 7 minutes ahead from the actual time. Her watch is ____ but not ___.

To test for reliability, a test group is split into smaller groups to see if they achieve similar results.

Ms. Smith saw that the distribution of scores on her last Bio test had an small standard deviation. This means that the.

When a test is repeated multiple times to see if the results remain the same it is known as split-half reliability.

The validity type that measures the extent to which test scores relate to future performance

in a normal distribution, 95% of the score should fall

New Zealand researcher known for publications on the increase in intelligence scores over time

The formula we use to calculate standard deviation is known as the _________ ________.

There is no concrete explanation on why the Flynn Effect occurs. We can only speculate.

Why do we need to re-standardize tests?

A scale that measures intelligence and cognitive ability in adults.

Admissions officers use your highschool transcript to see how well you will do in college. this is ___ validity.

Most of the scores depicted in a Bell Curve Graph fall near the ____ or middle of the graph.

_____ ____developed intelligence scales such as the Wechsler Adult Intelligence Scale

Reliability refers to the ___ of the results. reliable results can be reproduced under the same conditions.

Over time, tests need to be ____ or improved to be more difficult in order to keep up with the Flynn Effect

The type of validity that tells us whether a scale/test measures a construct fairly is ___ validity.

Pablo took the ACT 4 times and got scores of 29, 30, 29 & 30. Since his results were consistent the test has high _____.

When a test is ___, it means that it is taken, administered, scored and interpreted the same way for all test takers

When a test is measuring in the right construct, but is not doing so in a consistent manner.