lecture 5 notes exam 1

Question 1

conceptual variable/construct

Answer 1

the high-level abstract idea, e.g., happiness

Question 2

conceptual definition

Answer 2

Your definition of the construct: ex: subjective sense of well-being (definition of happiness)

Question 3

operationalization

Answer 3

ways the construct can be measured, e.g., asking people how happy they are from 1 to 10

Question 4

How could you operationalize the construct of anxiety?

Answer 4

measuring heart rate, looking at speech patterns, how much they are sweating, their sleeping patterns, asking someone to rank their anxiety on a scale

Question 5

3 common types of measures

Answer 5

• Self-report: ranking on a scale or asking someone how many anxious moments theyve had this week
• Observational measures (behavioral measures), e.g., stuttering during a conversation, how many hours they sleep, fingernail biting
• physiological measure: sweating, saliva (cortisol levels)

examples are related to measuring anxiety

Question 6

what does it mean if a measure is reliable?

Answer 6

it is consistent

Question 7

what are the types of reliability?

Answer 7

test retest
interrater
internal

Question 8

what is test-retest reliability?

Answer 8

consistent scores everytime the measure is used, ex: someone will have about the same score today tomorrow next week or next year
comparing time 1 and time 2

Question 9

what is interrater reliability?

Answer 9

consistent scores no matter who measures. A measure has high interrater reliability if 2 people using the measure get similar scores for the same participant
comparing rater 1 and rater 2

Question 10

what is internal reliability?

Answer 10

consistent scores on different versions of questions
a measure has high internal reliability if a participant provides a consistent pattern of responses regardless of how the researcher phrased the question
comparing question 1 to question 2

Question 11

How can we visualize reliability?

Answer 11

scatterplots (the closer the points are to the line, the more reliable) and correlation coefficients (more reliable than scatterplots)

Question 12

in what reliabilities are scatterplots used?

Answer 12

interrater reliability and test-retest reliability

Question 13

correlation coefficients (r)

Answer 13

ranges between -1 and 1
any r value greater than 0 is positive anything below 0 is negative
strong associations are closer to -1 or 1; weak associations are closer to 0 (test-retest above 0.5, interrater above 0.7)

Question 14

where would we expect high association vs low association?

Question 15

scales of measurement

Answer 15

nominal, ordinal, interval, ratio

Question 16

categorical/nominal scales

Answer 16

Categorical/nominal variables: levels are qualitatively distinct; order doesn’t matter
- e.g., social media apps: TikTok, Snapchat, Instagram,
Facebook.
- current TV shows on netflix

Question 17

ordinal scacles

Answer 17

Ordinal: ranked order
- Numbers corresponding to each level are meaningful.
- The distance between the levels doesn’t matter
Examples
- Place in a race: 1st, 2nd, 3rd. the distance between first
and second could be 10 seconds apart, and the distance
between second and third could be 4 seconds apart
- socioeconomic status

Question 18

interval scales

Answer 18

Interval scale: equal distance between levels
- numbers corresponding to each level are meaningful
- the distance between the levels DOES matter
- there is no true 0; a level of 0 doesnt mean theres no
variable present
Examples
- jeans sizes: size 0, 2, 4, 6 (a size 0 doesnt mean the
person has no waist
- the year you were born

Question 19

ratio scales

Answer 19

Ratio scales: equal distance between levels and 0 is meaningful

Examples:
- Asking someone how many cats they have. If someone has
0; they have none If someone has 4, they have 4
- number of classes you are taking

Question 20

inter-item reliability

Answer 20

usually 2 or more items in a survey so one correlation coefficient isnt enough

Question 21

validity of measurement

Answer 21

1. Face/content validity
2. Criterion validity
   3.

Question 22

face/content validity

Answer 22

Does it look like a good measure, subjectively?
face validity: it looks like what you want to measure, ex: a math test looks like a math test. high face validity
content validity: it contains all the parts that your theory says it should contain ex: the math test only contains addition and no other things like subtraction and division so low content validity

Question 23

criterion validity

Answer 23

What is your measure supposed to predict?
Known groups paradigm: examine whether scores on your
measure meaningfully differ between groups whose behavior is already well understood