TYPES OF MEASUREMENT SCALES:

I. nominal

Ex: Which of the following would you most like to have? (circle one)

Restored 1967 Mustang (=1)

Mazda Miata (=2)

Jeep Cherokee (=3)

Mercedes Sedan (=4)

Can measure frequencies with nominal scales


II. ordinal -- categories can be ordered in terms of magnitude

Ex: What is your hometown=s approximate population? (circle one)

less than 10,000

 10,000-50,000

50,001-150,000

150,001-500,000

500,000 or more

 

III. interval

Ex: How would you rate your eating habits?

         -2    -1     0     1     2
very unhealthy                 very healthy   

(Likert scale)

 

IV. ratio

Ex: How tall are you?

Has an absolute zero

 

WRITING METHODS SECTIONS:

Method

Participants

Number of participants, demographic info, what they got in exchange.

Design Levels of independent variable, what dependent variable was.

Materials

Stuff you used, measures, equipment.

Procedure

Step by step, everything you did, ideally so someone else could repeat the procedure by following what you have reported.


RELIABILITY AND VALIDITY:

Validity (also referred to as CONSTRUCT VALIDITY): Is your measure a good measure of the theoretical variable you are interested in?

Other special kinds of validity:

FACE validity: Does your measure on the surface seem related to the theoretical variable you are studying?

CONVERGENT validity: Does your measure correlate with theoretically related measures?

DISCRIMINANT validity: Is your measure unrelated to other measures that are designed to assess different theoretical variables?

reliability: Does a measure produce consistent results?

test/retest reliability - does a measure yield the same results when used on more than one separate occasion?

interrater/interobserver reliability - can two or more people agree in their observations or ratings?

split-half reliability - does one half of a measure (for example, a personality test) yield similar results to the other half?


SAMPLING:

population--all members of an identifiable group; the group to which we want to generalize our findings

sample--a subset of a population

element--one member of a population

representativeness--the extent to which a sample has the same distribution of characteristics as the population from which it was drawn.

(essential to generalize from sample to population)

 

selection bias vs response bias

 

probability sampling

simple random sampling

stratified sampling

cluster sampling

 

non-probability sampling

sample of convenience-

purposive sampling

 

 

OBSERVATION and SURVEYS:

naturalistic observation: JUST watch, no intervention

 

observation with intervention:

participant observation

structured observation

field experiments

 

external validity vs control:

external validity is the the extent to which the results of a study can be generalized beyond the context of the experiment, to the Areal world.@

confederate--Someone working with a researcher who is instructed to behave in a certain way in order to produce a particular condition.

expectancy - bias causing us see what we expect

Safeguards against expectancy:

1. good operational definitions

2. good recording format

3. interobserver (interrater) reliability

4. Ablind@ observers (observers who are unaware of the hypotheses)

 

subject biases--

Good subject

Demand characteristics

Reactance

Hawthorne effect

Evaluation apprehension


Social desirability bias - when people respond by trying to put themselves in a favorable light