Psychological Assessment (Diagnosis) pt.6

Coefficient Alpha

Received the most acceptance and is widely used today

Coefficient Alpha

Ranges from 0 to 1

Coefficient Alpha

Usually used if the items are non-dichotomous (no right or wrong answers).

A perfect example is the Likert Scale (e.g., measuring attitude or personality—choosing 5 does not mean you are "correct" because there are no right or wrong answers in a Likert Scale).

KR20

Although Coefficient Alpha is usually used for non-dichotomous or polytomous items, remember that it can also be used for dichotomous items. In situations where ___ can be used, Cronbach's Alpha may also be used, but the reverse is not true.

Coefficient Alpha

Although ____ is usually used for non-dichotomous or polytomous items, remember that it can also be used for dichotomous items. In situations where KR20 can be used, Cronbach's Alpha may also be used, but the reverse is not true.

Coefficient Alpha

Most commonly used.

Coefficient Alpha

The closer the ______ value is to 0, the lower the reliability. The closer it is to 1, the higher the reliability.

0

The closer the coefficient alpha value is to __, the lower the reliability. The closer it is to 1, the higher the reliability.

1

The closer the coefficient alpha value is to 0, the lower the reliability. The closer it is to __, the higher the reliability.

dichotomous items

Although Coefficient Alpha is usually used for non-dichotomous or polytomous items, remember that it can also be used for______. In situations where KR20 can be used, Cronbach's Alpha may also be used, but the reverse is not true.

Excellent

Coefficient Cronbach's Alpha: more than .90, it's reliability level is

Good

Coefficient Cronbach's Alpha: 0.80 - 0.89, it's reliability level is

Acceptable

Coefficient Cronbach's Alpha: 0.70 - 0.79, it's reliability level is

Questionable

Coefficient Cronbach's Alpha: 0.6 -.69, it's reliability level is

Poor

Coefficient Cronbach's Alpha: 0.5 - .59, it's reliability level is

Unacceptable

Coefficient Cronbach's Alpha: Less than 0.59, it's reliability level is

.90

Steiner (2003) pointed out that the value of the alpha above ___ may be too high and indicate redundancy in the items

too high

If Cronbach's Alpha is above .90, it might be ____.

There may be redundant items.

Example:

Item 1: I am down most of the time.

Item 6: I often feel down.

Item 3: I feel sad.

Item 4: I feel blue.

These statements are synonymous, so test-takers will likely give the same answers.

questionable

That's why a Cronbach's Alpha of .98 is ____ because it is too high, which may indicate redundancy. If redundancy exists, the test should be revised.

redundancy

That's why a Cronbach's Alpha of .98 is questionable because it is too high, which may indicate ______. If redundancy exists, the test should be revised.

.70

In research, a Cronbach's Alpha of ___ is already acceptable, meaning the test can be used for research after pilot testing.

.70

A ___ Cronbach's Alpha is acceptable for low-stakes tests, such as basic research, where results are not life-changing

.90

For high-stakes tests, a Cronbach's Alpha of ___ is required.

Inter-rater Reliability

Is the degree of agreement or consistency between 2 or more scorers with regard to a particular measure

Ex of behaviors scored by raters: depressed mood (as observed) or nonverbal performance task

Often used in coding nonverbal behavior

Inter-rater Reliability

Is the degree of agreement or consistency between 2 or more scorers with regard to a particular measure

Inter-rater Reliability

Often used in coding nonverbal behavior

Inter-rater Reliability

is usually used for rating scales that are not self-report—meaning the test-takers do not answer them.

Instead, test users (e.g., observers or judges) evaluate or rate a particular person based on a checklist.

Inter-rater Reliability

It measures the degree of agreement/consistency between the scores/ratings given by one rater and another rater.

4o

Variable

When choosing what kind of reliability should be used, there are lot of things to be consider one of which is the

Stable vs. Dynamic

Nature of a Test

Test-Retest Reliability

If the test measures a variable or a trait that is stable and we wish to assess the stability of the trait across different situations, then we use _____

stable

If the test measures a variable or a trait that is ____ and we wish to assess the stability of the trait across different situations, then we use Test-Retest Reliability

Test-Retest Reliability

Example: Stable - does not change significantly over time.

Most personality traits are often stable.

Example: If you are conscientious today, you will probably still be conscientious next month.

Intelligence is also relatively stable over time.

If a variable is stable, it would be wise to use ____ to determine if the test remains consistent over time.

Alternate Form Reliability

If the concern is still stability, but we would like to control for practice and carry over effect, and if a test has more than one form, then we use _____

Control for practice and carry-over effect

These could be sources of error.

Dynamic Variables

Usually applies to states rather than traits.

Example: Anxiety - You might feel anxious today, but two days later, you might not feel anxious anymore.

Test-Retest Reliability

If you use _____ to measure an anxiety scale, it might not be appropriate. Imagine testing anxiety levels in students before a board exam - their anxiety is likely high. Then, you test them after the board exam - their anxiety is likely low.

The results would show low consistency, making it seem like the anxiety test is unreliable, when in fact, the problem is using the wrong reliability measure.

Instead of _____, the better option would be Alternate Form Reliability.

low consistency

If you use Test-Retest Reliability to measure an anxiety scale, it might not be appropriate. Imagine testing anxiety levels in students before a board exam - their anxiety is likely high.

Then, you test them after the board exam - their anxiety is likely low.

The results would show ____, making it seem like the anxiety test is unreliable, when in fact, the problem is using the wrong reliability measure.

Instead of Test-Retest Reliability, the better option would be Alternate Form Reliability.

Alternate Form Reliability

If you use Test-Retest Reliability to measure an anxiety scale, it might not be appropriate. Imagine testing anxiety levels in students before a board exam - their anxiety is likely high.

Then, you test them after the board exam - their anxiety is likely low.

The results would show low consistency, making it seem like the anxiety test is unreliable, when in fact, the problem is using the wrong reliability measure.

Instead of Test-Retest Reliability, the better option would be ______

Kurtosis

This refers to the tailedness of a distribution.

Skewness

Extent of the absent of the symmetry

Normality

Extent of the presence of symmetry

Kurtosis

Is a measure of tailedness of a distribution - how often outliers occur?

It is not concerned with peakness nor variability but with the weight of the tails in comparison to a normal distribution (Weunch, 2014)

Kurtosis

Is a measure of tailedness of a distribution - how often outliers occur?

Kurtosis

It is not concerned with peakness nor variability but with the weight of the tails in comparison to a normal distribution (Weunch, 2014)

weight

Kurtosis is not concerned with peakness nor variability but with the ___ of the tails in comparison to a normal distribution (Weunch, 2014)

Kurtosis

Also used to describe the population, sample, or data set.

It does not refer to the steepness of the distribution.

It measures how frequently outliers or extreme values appear in a distribution.

It examines how heavy the tails are compared to the shoulders of the distribution.

Leptokurtic

When Gino checked the distribution of the scores of the test he made, he noticed that there were many outliers at both ends of the tail of his distribution. Thus, we could say that this might be a what kind of distribution?

Mesokurtic

Leptokurtic

Platykurtic

Three Types of Kurtosis

Mesokurtic

Medium tailed; outliers are neither highly frequent nor infrequent