5 Written Questions
5 Matching Questions
- on-site survey station
- inductive reasoning
- methodology bias
- casual inference
- nonprobability (non scientific) samples
- a self- administer response, can have texts and graphics, skipping/ no skip software. respondents are available, not true sample of pop. software can restrict those who participate
- b an agency automatically defaults to a method that they are most comfortable with
- c drawing a conclusion about a cause and effect relationship based on observation of some occurrence and the conditions that existed at the time (the cause)
- d makes general conclusions based on individual observation
- e each member does not have a known chance of being selected, researcher randomly chooses sampling units. inexpensive and easy. used for questionnaire testing and prelime stages
5 Multiple Choice Questions
- similarities between two things to a conclusion about some additional attribution of both things.
- 2 types: unstructured in home and structured intercepts.
- premise based on a sample group and leads to a conclusion involved another individual- group of customers to the next customer you see
- list of the people you select for your sample
- like stratified sampling, not random. most common form of nonprobability. must complete a quota for each strata- like 40 blacks, etc.
5 True/False Questions
cost per return → showing a series of stimulu in a specific order and not rotating the elements to avoid bias in the manner in which they were shown. sequence in which ads shown
generalization → future conclusion from some sample in the past
electronic surveys → email, web based, on cite stations. samples do not typically represent target population- usually from email list. internet most frequent in urban and higher income and education. inexpensive and quick to collect info. self administered, must be simple
external validity → did the research project find out what it said it was going to? ex. if participants lie
web- based survey → database and html, can skip to certain questions based on answer and wont let respondent skip. can provide animation. more complicated to design than email. fast, low cost, flexible. population sample limitation