A. Inductive. "Context of discovery." Interested in how and what (idiographic--here, specific), rather than why.

1. Collect information on events--episodes, happenings.

2. Identify "family resemblance's" (Wittgenstein, Philosophical investigations)--similar features among events. Create categories, classes, ideas. Define each class by common, distinguishing features; e.g., "These events have in common an individual's effort to manage other persons' impressions of the individual as one of them, valuable, 'normal.'" The concept is an empirical (derived from experience) concept. Give it a name: "normification."

3. Collect more events to determine the range of events in the concepts--the limits. Perhaps create typologies; e.g., "There are seven kinds of normification."

4. Look for relationships. For example, women tend to normify using methods 1-3 (categorical). Generally the first method of normification is... Or, the highest rates of resident abuse occur during times when staff have a lot of tasks to accomplish in little time and the ratio of staff to residents is low. State relationships as empirical empirical generalizations.

Rules of inductive inference are used to tie alleged causes or antecedents (independent variables) to alleged effects (dependent variables):

a) alleged causes precede alleged effects;

b) alleged causes and alleged effects are both observed; If relationships are distal, the intervening steps are observed;

c) Mill's methods are used to make logical inferences: concomitant variation; method of agreement, method of difference, joint method of agreement and difference, method of residues;

d) plausible rival explanations are ruled out.

B. Deductive. "Context of verification." Interested in why (and in general principles/laws/regularities--nomothetic).

1. Assert causal/functional propositions/hypotheses to test/verify/falsify-- which could be empirical generalizations (step 4) from inductive research. E.g., "Social deviance varies inversely with social cohesion."

2. Define each conceptual variable (social deviance, social cohesion).

3. Deduce operational variables/definitions from the conceptual definitions. E.g., rates of suicide, homicide and mental illness (deviance) vs percentage of population regularly attending church, employment rate, rates of divorce (examples of social cohesion). Assert expected relationships among operationalized variables; e.g., "The higher the unemployment rate (independent variable), the higher the rates of mental illness, suicide, and homicide (dependent variables)." These are operational hypotheses.

4. Collect data on operational variables (e.g., official statistics) and use methods of inductive inference to verify/falsify hypothesized relationships among operational variables.

5. Suggest implications of findings for the original hypotheses and for further research; e.g., replication.