Textbook Notes (368,794)
Psychology (9,697)
PSYB01H3 (581)
Anna Nagy (283)
Chapter 5

# Chapter 5 - Measurement Concepts.docx

3 Pages
81 Views

Department
Psychology
Course
PSYB01H3
Professor
Anna Nagy
Semester
Winter

Description
Chapter 5: Measurement Concepts Reliability of Measures Reliability: the consistency or stability of a measure of behaviour - reliable measures do not flucuate form one reading to another - any measure can be comprised of two components o true score: the real score on the variable o measurement error: the degree to which a measurement deviates from the true score value [reliable sores have little measurement error] - in research we measure each person only one so important to use reliable measure - cannot use unreliable measure to systematically study variables or relationships - reliability can be increased by making multiple measures [common: personality] o more items the higher the reliability - cannot directly observe true score and error but can assess stability of measures using correlation coefficients - Pearson product-moment correlation coefficient: r, -1.00 to +1.00 o Reliability correlation - To assess reliability we need at least 2 scores from many individuals Test-Retest Reliability: assessed by measuring the same individuals at two points in time - reliability coefficient should be at least .80 - ie. measure group one day and another day of the week - may be high because ppl rem how they answered before can use alternative forms - some very reliable – IQ, some not – mood Internal Consistency Reliability: the assessment of reliability using responses at only one point in time [several items assessing one measure] - split-half reliability: the correlation of an individual’s total score on one half of the test with the total score on the other half of the test o the combined measure will have more items and will be more reliable than either half by itself o does not take into account each individ item’s role in a measure’s reliability - Cronbach’s Alpha: based on the individual items o The correlation of each item with every other item - Item-total correlation: examine the correlation of each item score with the total score based on all items o Items that do not correlate with others can be eliminated > increasing reliabi Interrater Reliability: the extent to which raters agree in their observations [Cohen’s kappa], high number of observations that result in similar judgment Reliability and Accuracy of Measures - example: shoe size measurement device as intelligence measure - difference between reliability and accuracy of measure! Construct Validity of Measures - construct validity: refers to the adequacy of the operational definition of variable - does the measure employed measure actually measure the construct intended Indicators of Construct Validity Face Validity: the content of the measure appears to reflect the construct being measured - no
More Less

Related notes for PSYB01H3
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.