Term
|
Definition
|
|
Term
|
Definition
| we draw a sample of households, the households are the enumerated sample and the individuals are the sample |
|
|
Term
|
Definition
| the list from which the elements of a popluation are selected |
|
|
Term
|
Definition
| refers to the ability to generalize from a sample or subset of a larger population to that population itself |
|
|
Term
| cross-population generalizability |
|
Definition
| refers to the ability to generalize from one group, population to other groups or populations |
|
|
Term
|
Definition
| refers to only to cross population generizabily, NOT to sample generalizabitly |
|
|
Term
|
Definition
| any difference btwn the characteristics of a sample and the characteristics from which it was drawn |
|
|
Term
|
Definition
| is a set of the elements larger than or different from the pop. that was sampled and to which the reseacher would like to generalize any study findings |
|
|
Term
|
Definition
| good measure of the target pop, and balaced the sample, with similar makeup as the whole. |
|
|
Term
| probability sampling methods |
|
Definition
| sampling methods that allow us to know in advance how likely it is that any element of a pop will be selected for the sample |
|
|
Term
| Non-probabiblty sampling methods |
|
Definition
| sampling methods that DO NOT let us know the likleyhood in advance that any element of a population will be selected. |
|
|
Term
|
Definition
| because ppl who do not respond are likely to differ systematically from those who take the time to participate |
|
|
Term
|
Definition
| requires a method that generates numbers or cases only on chance. |
|
|
Term
|
Definition
| easy way of random selection. can be obtained on many websites. |
|
|
Term
|
Definition
| an automated machine dials a phone number randomly |
|
|
Term
|
Definition
| when the number sample is replaced into the sampling frame after it has just been selected. |
|
|
Term
| who used random simple sampling by telephone and had similar results to the us census data? |
|
Definition
| Tjaden and Thoennes (2000) |
|
|
Term
| systematic random sampling |
|
Definition
| the starting point is selected randomly then you only pick every Nth number when the files are sequential |
|
|
Term
|
Definition
| part of systematic random sampling... EX every 20th case is selected. 20 is the sampling interval. |
|
|
Term
|
Definition
| staight-forward, element are selected because they are easy to find. AKA convienence sampling. News ppl interviewing ppl on the steet. whoever happends to be around at the time. |
|
|
Term
|
Definition
EX. 10% of pop must be over 65yrs old
50% under 30yrs old etc. tries to overcome limitations of avalibilty sampling. |
|
|
Term
|
Definition
| each sample element is selected for a purpose. EX studying the population of some limited group like a street gang-- not random |
|
|
Term
|
Definition
| talk to one person, then offer incentive to get him to give up name of others to talk to. works well for hard to reach populations like gangs |
|
|
Term
| who used snowball sampling in a Chicago neighborhood? |
|
Definition
|
|
Term
|
Definition
| a hypothetical destribution of a statistic across an infinate number of random samples that could be drawn from a population |
|
|
Term
| what does normal distrubution look like? |
|
Definition
| looks like a bell shape, centered around the population mean and both sides taper of symetrically. |
|
|
Term
| what is symetry the normal distribution graph produced by? |
|
Definition
|
|
Term
| systematic sampling error |
|
Definition
| overrepresentation or underrepresentation of popluation characteristics due a bad method of sampling. |
|
|
Term
|
Definition
| overrepresentation or underrepresentation of popluation characteristics due pure chance. |
|
|
Term
|
Definition
| is an estimate of the pop. parameter we want to estimate. |
|
|
Term
|
Definition
| the statistic computed ofr the entire population. |
|
|
Term
|
Definition
| the inside of the upper and lower bounds of the graph. |
|
|
Term
|
Definition
| upper and lower bounds of the confidence interval. |
|
|
Term
|
Definition
| units under study... groups, towns, individuals, etc. |
|
|
Term
|
Definition
| data collected from individuals. sometimes the same as units of analyis. |
|
|
Term
| when a researcher draws a conclusion about an individual from the group data is making a ________. |
|
Definition
|
|
Term
| when a researcher draws a conclusion about a group from and individual he is making an ___________. |
|
Definition
|
|
Term
| nonothetic causal explanation |
|
Definition
| the belief that variation in an independent variable will be followed by variation in the dependent variable. |
|
|
Term
| the situation that would be in absense of any change in the independent variable |
|
Definition
|
|
Term
| idiographic causal explanation |
|
Definition
| EX: an individual is neglected by his parents, then he distrusts others, then has trouble in school, then gets addicted to herion. AKA causal effect |
|
|
Term
| who did the experiment with watching the violent video tapes? |
|
Definition
|
|
Term
| non-spurious relationship |
|
Definition
| a relationship that is btwn the two variable being studied, NOT by any third variable |
|
|
Term
|
Definition
| the third variable that causes a relationship to be spurious. |
|
|
Term
|
Definition
| when the third variable (extraneous variable) has an affect on the relationship, and no conclusion can be made btwn the two original variables that were studied. |
|
|
Term
|
Definition
| some way of creating a connection btwn two the two variable being studied. |
|
|
Term
| true experiments must have what three things? |
|
Definition
Two comparison groups, one recieving treatment, and one w.out
Random assignment of the two groups
Assesment of change in the dependent variable for both groups afterwards. |
|
|
Term
| non-experimental designs can be which two kinds? |
|
Definition
| longitudinal or cross-sectional |
|
|
Term
|
Definition
| all data are collected at one point in time. |
|
|
Term
| longitudinal research designs |
|
Definition
| data are collected at more than one point in time. |
|
|
Term
| who used the Arrestee Drug Abuse Monitoring (ADAM)to study offenders incararated in Ohio county jails? what did they use to help the respondants recall the info in correct time-order |
|
Definition
| Lo, Kim, Cheng (2008)... anchors |
|
|
Term
| repeated cross-sectional design |
|
Definition
| AKA trend study. type of longitudinal study that allows more people to enter the same population that was studied previously. |
|
|
Term
| fixed-sample panel design |
|
Definition
| type of longitudinal study where data is collected from the same individuals at 2 or more points in time. |
|
|
Term
| event-based design. definition and what is it AKA? |
|
Definition
| AKA corhort study. A longitudinal study in which data are collected at two or more points in time from people from the same cohort. |
|
|
Term
|
Definition
| people who have all experienced common evert or starting point. EX: all ppl born in 1950's |
|
|
Term
|
Definition
| technique in non-experimental research to reduce spuriousness. |
|
|
Term
| from nomothic perspective: what are the 3 criteria for a causal relationship? |
|
Definition
| Associaton btwn the variable, proper time order, and non-spuriousness of the association. |
|
|
Term
| true or false? "Correlation does not prove causation" |
|
Definition
|
|
Term
| Why are are idiographic causal explainations difficult to explain? |
|
Definition
| because the starting and ending points and which events act as causes in certain sequences make it ambiguous or hard to distinguish. |
|
|
Term
| Which longitudinal design provides the strongest test for time-order of effects? |
|
Definition
| Fixed-sample panel designs, but they are hard to do b/c they are expensive and wear mentally on those who participate in the studies. |
|
|
Term
| Who drove down streets in Chicago and recorded the level of disorder the observed? |
|
Definition
| Sampson and Raudenbush (1999) |
|
|
Term
| treatment is considered which variable? |
|
Definition
|
|
Term
| true or false? all true expirements have a posttest as well as a pretest. |
|
Definition
| False. All have a posttest, some have a pretest. |
|
|
Term
|
Definition
| measure the dependent variable prior to the experiment. |
|
|
Term
|
Definition
| all true experiments have this. Measures the dependent variable affect the treatment has been given. |
|
|
Term
| the best alternative to an experimental design is usually considered __________. |
|
Definition
| quasi-experimental design |
|
|
Term
| what are the 3 types of Quasi-experimental designs? |
|
Definition
| Non-equivalent control group design-have experimental and comparison grous designated before the treatment and are not randomly assigned. Before and After designs- have pre and post test but no comparison group. Ex post Facto control group design- like non-equivalent groups but have a control group. |
|
|
Term
|
Definition
| people are matched to someone else with similar traits. |
|
|
Term
|
Definition
| matching group trends rather than indiviual ones. |
|
|
Term
| who studied the effectiveness of a drug court on recidivsm? |
|
Definition
|
|
Term
|
Definition
| deselection; groups become more different for various reasons and subjects become more likely to drop out of groups... common in field experiments |
|
|
Term
|
Definition
| occurs when natural things happen to people in the study, not related to the treatment, that acount for some or all of the observed changes btwn the pretest and the posttest. |
|
|
Term
| what are the three threats to internal validatiy for endogenous change? |
|
Definition
| Testing, Maturation, Regression |
|
|
Term
|
Definition
| subjects may age, gain knowlege in school, etc and respond differently in the posttest than they did in the pretest. |
|
|
Term
|
Definition
| ppl in a study who do badly in the pretest may have had a bad day and then consequently do better on the posttest when they take it on a better day. |
|
|
Term
|
Definition
| AKA history effect. Things that happen outside the experiment that affect the outcome. |
|
|
Term
|
Definition
| when the control group is affected by the treatment group |
|
|
Term
|
Definition
| AKA John Henry effect. In honor of the steel driver who died trying to compete with the steam drill, when he learned he was being compared to it. |
|
|
Term
| treatment misidentification |
|
Definition
| occurs when the treatment does not cause the outcome, and the researcher does not pick up on it. |
|
|
Term
| expectancies of the experimental staff |
|
Definition
| may cause treatment misidentificaiton |
|
|
Term
|
Definition
| moved workers to differently lite areas to determine productivity. they found that only the fact that they knew they were being studied affected the productivity. |
|
|
Term
|
Definition
| periodic measures that are taken throughout the experiment to assess whether the treatment is being delivered as planned. |
|
|
Term
|
Definition
| respondents are asked about their likley responses to hypothetical situations to maximize survey generizablity. |
|
|
Term
| which variable is represented by treatment or intervention? |
|
Definition
|
|
Term
|
Definition
| involves the collection of information from a sample of individuals through their response questions. |
|
|
Term
|
Definition
| covery many different topics to social scientists. one of most successful is the General Social Survey (GGS). |
|
|
Term
|
Definition
| after the open-ended question has been asked, the researcher askes an additional research question |
|
|
Term
| another word for close-ended questions |
|
Definition
|
|
Term
| double negative question example: |
|
Definition
| "Do you disagree that juveniles who commit murder should be tried as an adult?" |
|
|
Term
| doubled barreled question |
|
Definition
| asking two questions with one statement. |
|
|
Term
| mutually exclusive responses |
|
Definition
| when there is no overlap in the possible response choices |
|
|
Term
|
Definition
| generally ask for participants to how strongly they are for or against the question. |
|
|
Term
|
Definition
| people who see themselves as neutral on a subject. |
|
|
Term
|
Definition
| ppl that choose a question even when they do not know about what the question is asking. |
|
|
Term
|
Definition
| to make sure the question you are asking only is asked to a relevant responder |
|
|
Term
|
Definition
| when a participant is asked a question and the response dictates that the not be asked the next question in series. |
|
|
Term
|
Definition
| part of a skip pattern that the participant will only be asked the question if you answers "yes" to the first question. This second question related to the filter question is the contingent question. |
|
|
Term
| idiosyncratic variation, which type of questions are prone to this? |
|
Definition
| single questions are prone; when a participant answers the question the way he does b/c of particular words or ideas in the question itself. |
|
|
Term
|
Definition
| -several questions are asked to measure one concept, the the measure is based on the sum or average of the answers. |
|
|
Term
|
Definition
| help researchers understand whether the responses are consistant or not. |
|
|
Term
|
Definition
| when you answer the question yourself and revise it. |
|
|
Term
|
Definition
| ask ppl test questions and then ask them follow-up questions to find out how good their test questions were. |
|
|
Term
| a cover letter should be: |
|
Definition
| mailed to repsondents and have these qualities: credible, personalized, interesting, responsible |
|
|
Term
| what makes a group-administered survey hard to do |
|
Definition
| assembling the group, ppl going about their daily activities do not count as this |
|
|
Term
| CATI and whats the advantage of this |
|
Definition
| computer assisted telephone interview, data collection can happen at the same time. |
|
|
Term
|
Definition
| interactive voice response, can also ask open-ended questions to be heard later. |
|
|
Term
| which survey type is the best if money is no object |
|
Definition
|
|
Term
|
Definition
| computer assisted personal interview, interviewers carry a laptop and enter in the responses they participants give. |
|
|
Term
| CASI- what is it good for? |
|
Definition
| computer assisted self-interviewing--good for sensitive information like victimization |
|
|
Term
|
Definition
| survey may be mailed to someone that doesn't have email, but emailed to someone that does have it. |
|
|
