Nonexperimental Methods II
Lecture, Chapter 5
Surveys and Questionnaires
Ex post facto study – “after the fact” study, involving no manipulation of the IV.
Types of Surveys
Methods of gathering data
Writing Survey Questions
Be clear, concise, and speak in participants’ language.
Avoid confusing phrasing, such as double negatives and double-barreled questions (asking 2 questions but allowing only 1 answer).
Minimize bias by avoiding emotional language, allowing disagreement, and presenting all possible choices.
Participants must be competent to answer questions; don’t ask questions they can’t answer.
Allow for uncertainty to prevent “floaters,” who really do not know, from contaminating results.
Make responses mutually exclusive (no overlap in choices) and exhaustive (all options are offered).
Types of Tests and Inventories
Characteristics of Good Tests and Inventories
lPopulation – the comprehensive group in question
lSample – small group selected to represent the population
lRandom sample – every member of the population has equal likelihood of selection
lWithout replacement (cannot be selected again)
lWith replacement (can be selected again)
lStratified random sample – random samples drawn from specific subpopulations or “strata” of the population
lSystematic random sample – every nth participant is selected
lCluster sampling – a naturally occurring, mixed aggregate of elements in a population, with each element appearing in 1 and only 1 cluster (ex. Restaurants can serve as clusters for sampling waitors.)
Nonrandom sample – members have different likelihoods of selection; useful in qualitative studies, when research question calls for intensive investigation of small population, or for preliminary, exploratory study.
lAvailability/convenience – using whoever is available (ex. Standing on the street corner)
lSnowball – identify 1 member of the population, ask that person to identify others, contact them, and so on.
lPurposive – each sample element is selected for a purpose, because of the unique position, until completeness and saturation is reached
lQuota sampling – availability sampling with certain quotas that represent the population (ex. Proportion of ethnic/cultural background of students in a university)
Basic Research Strategies