How would you calculate whether or not your test was reliable?
What will be an ideal response?
There are two main ways of calculating internal consistency reliability: split half reliability and coefficient alpha. Split half reliability works as follows: say we have an attitude to teaching measure that consists of 10 items. First, we randomly split the test into two (for example, the even and uneven items). Then we calculate respondents’ scores on each ‘half test’. We can then see whether the two scores are related to one another. If
they are both measuring the same thing, we would expect them to be strongly related, with a correlation coefficient of over .8. Coefficient alpha is another measure of internal
consistency. This measure would be expected to be over .7 before we can say that our test is internally consistent.
You might also like to view...
When is the optimal time when a person should intervene with a "student fighter"?
a. when the fight is occurring b. when the "student fighter" starts teasing another student c. when the fight is about to occur d. both b and c e. none of the above
In the name of equality and justice, Alabama and other southern states sought to provide school funding that was
A. apportioned to black schools and white schools according to the number of students enrolled. B. apportioned in favor of white schools. C. roughly equivalent for both types of schools. D. All these answers are correct.
How can the Flynn Effect impact the costs of standardized testing?
a. It has no effect. b. by requiring frequent revision c. by reducing demand d. by eliminating bias
The traditional board member was a white college-educated male ranging in age from 40 to 50 years old
Indicate whether this statement is true or false.