Intro to Research Chapter 5 Week 3 Lecture
Intro to Research Chapter 5 Week 3 Lecture
Popular in Intro to Research Methods
Popular in Department
This 5 page Class Notes was uploaded by Kim Notetaker on Friday September 30, 2016. The Class Notes belongs to at Armstrong State University taught by in Fall 2016. Since its upload, it has received 5 views.
Reviews for Intro to Research Chapter 5 Week 3 Lecture
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/30/16
Chapter 5 Scales of measurement (NOIR) Nominal: values are just labels and categories. o An example would be, are you happy? Yes or no. Ordinal Scale: the rank ordering and the amount of space between the ratings doesn’t matter. Interval: the equal differences between the numbers reflects the equal differences on the dimension being measured. This has NO true zero or absolute zer0. o An example would be how happy you are on a scale of 1-7. 1 being the least happy and 7 being the most. Ratio: the interval scales plus a ‘true zero.’ This is a good measure of behavior. o An example would be asking how many times did you feel happy today? How do we know if our measures are good? Validity: the degree to which a measure is an accurate representation of the construct we want to measure. Reliability: the consistency, stability or dependability of a measure. Correlation Refresher We typically measure reliability with correlation. Correlations Co-efficient (r): measures the strength and direction of the association between 2 variables. Strength: how well can you predict one thing by knowing about the other? (ranges from 0 to +/- 1) o 0 = no correlation o .1 - .3 = small correlation o .3 - .5 = medium correlation o .5 or larger = large correlation Direction: as one variable goes up does the other variable go up or down? Positive Correlation: all the variable either go up or down at the same time. Negative Correlation: one variable goes up while the other goes down. 3 assessments of reliability Test re-test Reliability: the consistency of the results over time. Relevant to ALL types of measures (self-report, observational and psychological) BUT only when the construct is stable. It measures more than once and calculates the correlation between the scores. o An example would be trait level happiness. Interrater Reliability: consistency over observers. Relevant to observational measures (video/Audio). Ask more than one person to rate the observations on a measure of interest, correlate the two sets of rating. o An example would be observed happiness. Internal Reliability: consistency across scale items. 2 Relevant to self-report measures that have multiple items assessing the same construct. o An example would be the Subjective Happiness Scales by Lyubomirsky. Typically assessed with a type of correlation: Cronbach’s alpha. o Step 1: Compute all possible connections between the items. (item 1 with item 2 and so forth.) o Step 2: Take the average of these correlations. o Step 3: Fancy math with the average of the correlation and the number of items = Cronbach’s alpha. o You want it to be greater than .70. 5 Types of Construct Validity Construct Validity: How well did the researcher operationalize each variable? Face Validity: Does the measure seem like a plausible one, given the construct of interest? o Entirely subjective. (don’t over think this one!!) o Usually determined by experts. Content Validity: Does the measure capture all parts of the construct of interest? o Entirely subjective. o Usually determined by experts. Criterion Validity: Is the measure related to relevant objective outcomes? 3 Correlation between scores on your measure and concrete objective outcomes (no other subjective outcomes). o An example question to ask would be: What objective outcomes should happiness be related to? Another option is the know groups paradigm: this is when you take groups you know to be different and give them the measures. Convergent Validity: Is the measure related to other measures that assess similar constructs? Correlations between scores on your measure and scores on other subjective related measures. o A question to ask is: What other measures should happiness be related to? Satisfaction with life scale (Diener) Positive and negative affect schedule (Watson and Clark) Discriminant Validity: Is the measure NOT related to other measures that assess different constructs? o NOTE: Looking for weak correlations not negative ones. o Only important for constructs that are similar but distinct. 4 5
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'