Exam #2 Review Sheet (Chapters 6-10)Causation and Experimental Design - Chapter 6- Experimental Designso 3 common elements 1. At least 2 comparison groups (experimental & control) 2. Random assignment (solves a lot of problems in experimental designs; creates equal groups & require large samples; no characteristic which places individual in one group over the other)- Roughly same age, gender, major, etc. 3. Stimulus/change- applying some variable that you’ve changed or application of a program, etc.- 5 criteria of causationo Cause- an explanation for some characteristic, attitude, or behavior of groups, individuals, or other entities (such as families, organizations, or cities) or for events.o Causal effect- When variation in one phenomenon, an independent variable, leads to or results, on average, in variation in another phenomenon, the dependent variableo 1) Empirical Associationo 2) Appropriate time ordero 3) Non-spuriousness- A relationship between 2 variables that is not due to variation in a 3rd variable; variation in an IV causes variation in a DV; A criterion for establishing a causal relationship between two variables; when a relationship between two variables is not due to variation in a third variable Spurious relationship- A relationship between two variables that is dueto variation in a third variableo 4) Identifying a causal mechanism- Many social scientist believe that a causal explanation is not adequate until a causal mechanism is identified; What process or mechanism is responsible for the relationship between the IV & DV Mechanism- Some discernible means of creating a connection; A discernible process that creates a causal connection between two variableso Time order- Time order- A criterion for establishing a causal relationship between two variables. The variation in the independent variable must occur before the variation in the dependent variableo 5) Specifying the context in which the effect occurs Context- A focus on idiographic causal explanation. A particular out-come is understood as part of a larger set of interrelated circumstances. Contextual Effect- Relationships among variables that vary among geographic units or other social settings- Cross-sectional vs. longitudinal designso Cross-sectional- study in which data are collected at only one point in time; “snap-shot”; most common form of analysis; simplest & cheapest, but cannotfully capture change (implying causal direction but can’t say for sure findings are consistent over time)o Longitudinal- A study in which data are collected that can be ordered in time; also defined as research in which data are collected at two or more points in time; measures characteristics at multiple time points; more powerful for measuring change; more costly & difficulto Association- Necessary criterion for establishing a causal effect; A change in the independent variable is correlated with a change in the dependent variable;if there is no association between the 2 variables, there cannot be a causal relationship o Cohort- People who all have experienced a similar event or common starting point Event based design (cohort study)- follow up samples are collected (at one or more times) from the same cohort Birth cohorts-Those who share a common period of birth (ex. Those born in the 1950s) Seniority cohorts-Those who have worked at the same place for about 5 yrs., about 10 yrs., and so on School cohorts- Freshman, Sophomores, Juniors, & Seniors o Cohort Study aka event-based design (type of panel study in criminology)- A type of longitudinal study in which data are collected at two or more points in time from individuals in a cohort Panel study but w/ a group of people that shared some experience at the same time o Case study- following one single person over time (or very few cases); qualitative research; direct observation; spending a lot of time w/one person; Wholistic view of each case; incorporates time but not in the systematic precise timed way as quantitative research - True experiments- An experiment in which subjects are assigned randomly to an experimental group that receives the treatment or other manipulation of the independent variable and a comparison group that does not receive the treatment. Outcomes are measured in a post-test.o Elements of a true experiment Has an experimental group, control group, & is done over time (compare both post tests then if we have a pre-test, we look at it there was a change in both groups & why); Without a pretest, no way of knowing what caused the change; rate of change (sees which group changed more)o Pre-test and post-test Pre-test: Same as the post-test but administered at an earlier time; A true experiment does not require a pre-test; In experimental research, this is the measurement of an outcome (dependent) variable prior to anexperimental intervention or treatment (independent variable)- Examines differences in groups, not necessary but desirable (don't really need it because they should be equal to start w/if correctly randomly assigned) but we use them to measurechange, tests randomness, and examines conditions under which the stimulus had an impact (whether certain kinds of people responded different to the stimulus) Post-test: In experimental research, this is the measurement of an outcome (dependent) variable after an experimental intervention or treatment (independent variable)- after intervention, measures differences;o Experimental vs. control group Experimental- In an experiment, the group of subjects that receives thetreatment or experimental manipulation Control- In an experiment or study, a comparison group that receives no treatmento Random assignment (randomization)- A procedure by which each experimental subject is placed in a group randomly Useful for ensuring internal validity not generalizability Matching- A procedure for equating the characteristics of individuals in different comparison groups in an experiment- Problem- poor replacement for randomization; individuals can be matched only on a few characteristics so unmatched & unknown differences between the experimental & control group may influence outcomes Ceteris Paribus- Latin phrase meaning “other things being equal”- Quasi-experimental designs- A research design in which there is a comparison group that is comparable to the experimental group in critical ways, but subjects are not randomly assigned to the
View Full Document
Unlocking...