chapter 6 RM

Observational research

naturalistic observation and participant observation

naturalistic observation

knowledgable informents

unobstrusive measures

participant observation

Advantages: You get rich, real-world data because you see behaviors as they naturally happen.

Disadvantages: Researcher bias is high — your presence and personal perspective can affect what you notice and how you interpret it.

Observer bias 

pre-existing expectations or beliefs

Open-end questions 

where they give you more than just yes and no question 

forced-choice questions

where they give you questions that are only limited and can’t give further explanation

like-scale questions

a type of survey question that asks respondents to rate their level of agreement or disagreement with a statement 

How to write good questions: general guidelines

Use langauage appropriate for population 

use precise wording 

aviod biased words 

aviod double barrled questions 

aviod unwarrented assumptions 

do not as questions that respondents may not know the anwser for 

do not ask questions for whcih everyone gives the same anwser 

Sampling goal

accurate description of population more for descriptive research and experimental research 

research dilemma

need a meausre of entire group (=population)

cannot measure each individual 

representative sample

sample that affords accurate and unbiased estimates about the population as a whole

biased sample

sample that does not afford accurate and unbiased estimates about the population as a whole 

central limit theorem 

large sample size the sample mean will reach a sample distribution

sampling error 

difference between population mean and sample mean 

does not respect other sources of error 

only meaning for probability sampling methods 

simple random assignment sampling 

equal chance of being selected from population 

stratified random sampling

divide the population into non-overlapping groups

do simple random sampling of each group

convenience (availability) sampling

recruit readily available research participants

psychology departments pool, students in class, patients at local hospitals 

not representative of population 

quota sampling

A convenience sample that ensures a certain distribution of demographic variables

Set goals of key demographic variables 

recruit participants until goals are met: same problems as a convenience sample 

snowball sampling

asking participants to recommend others for the study (good for targeting rare populations)

non-representative because social connections are not random