Methods I

Alla Inga Spara

• Aggregate a quantity that is created by combining the values of many individual cases.
• Replication a scientific process in which researchers implement the same procedures repeatedly in identical form to see if the relationships hold in a consistent fashion.
• Sample of convenience a sample of cases from the underlying population in which the mechanism for selecting cases is not random.
• Survey experiment a survey research technique in which the interviewing process includes experimental randomization in the survey stimulus.
• Confounder an outside variable that affects both the independent and dependent variables, potentially distorting the relationship between them and leading to incorrect conclusions about cause and effect.
• Natural experiment a method where real-world events or processes create conditions similar to a true experiment, with causes assigned randomly or nearly randomly, allowing for comparison and causal conclusions.
• Stratification the process of dividing a population into subgroups based on specific characteristics to improve the accuracy of comparisons and control for confounding variables.
• Omitted variable bias this occurs when a model leaves out a relevant variable that influences both the independent and dependent variables, leading to incorrect estimates of the relationship between them.
• Impact evaluation an assessment of changes in the well-being of individuals that can be attributed to a particular project, program, or policy.
• Descriptive questions ask about what is taking place, focusing on processes, conditions, organizational relationships, and stakeholder views.
• Mixed Methods evaluations that integrate both qualitative and quantitative analysis.
• The counterfactual problem the issue of determining what would have happened if an alternate scenario had occurred, such as the absence of a program or intervention.
• Cross-sectional measure a measure where the time dimension is the same for all cases, and cases represent multiple spatial units.
• Incomplete information a situation in which each actor in a game does not know the exact payoffs from each possible outcome.
• Intransitive relationship an illogical relationship where, despite A > B and B > C, C > A.
• Rational choice application of game theory and formal mathematical tools to puzzles of human behavior (also known as formal theory).
• Spatial dimension the physical units on which a variable is measured.
• Time dimension the point or points in time at which a variable is measured.
• Time-series measure a measure where the spatial dimension is the same for all cases, representing multiple time units.
• Transitive a logical relationship where, if A > B and B > C, then A > C.
• Utility a calculation equal to the sum of all benefits minus the sum of all costs from an action.
• Case a spatially delimited phenomenon (unit) observed at a single point in time or over a period.
• Case study method of intensive study of a single case to shed light on a larger class of cases.
• Cross-case study examines multiple cases to find patterns and generalize findings.
• Sample the cases of a larger population that are subjected to formal analysis.
• Least-Likely Case a case not predicted to achieve a particular outcome, used to confirm or strengthen a theory if the outcome occurs.
• Most-Likely Case a case expected to achieve a certain outcome, used to disconfirm or weaken a theory if the outcome does not occur.
• Most-Similar Systems Design compares cases that are alike in all aspects except the variables being studied.
• Most-Different Systems Design identifies cases where just one independent variable and the dependent variable covary, while all other factors differ.
• Scope conditions the theory’s domain—the set of cases over which a theory is expected to operate.
• Inductive theorizing explanation emerges from specific cases and is generalized outward.
• Deductive theorizing explanations emerge from abstract analysis, not based on specific cases.
• Hypothesis a statement of the expected relationship between the dependent and independent variables.
• Falsifiers evidence expected if a hypothesis is incorrect.
• Validity the accuracy of matching indicators with concepts.
• Process Tracing method that examines evidence to explore causal mechanisms within a case through empirical tests.
• Structured Focused Comparison applies the same theoretical questions across cases to enable systematic comparison.
• Method of Congruence examines if outcomes align with theoretical predictions, often complementing process tracing.
• Measurement the process of converting information into comparable, systematized values of specific variables for observations.
• Nominal categorical data without a specific order (e.g., gender, race).
• Ordinal categorical data with a specific order but no consistent difference between values (e.g., rankings).
• Ratio values that can be ranked and ordered, with equal differences and a natural zero (e.g., income).
• Interval values that can be ranked and ordered with equal differences but no natural zero (e.g., temperature).
• Reliability the consistency of a measurement tool in returning the same value when applied repeatedly.
• Counterfactual a hypothetical scenario for evaluating what could happen if a variable or condition were different.
• Elite Interviewing qualitative research method involving interviews with influential individuals for insights on specialized subjects.
• Snowball Sample a non-probability sampling method where study subjects recruit future subjects from their acquaintances.
• Code converting raw information into data using specific rules and categories to establish variable values.
• Categorical variables variables with values that differ without universally holding ranking distinctions.
• Central tendency typical values for a variable at the center of its distribution.
• Continuous variable a variable with equal unit differences where each increase signifies the same change across values.
• Dispersion the spread or range of values of a variable.
• Face validity whether a measure appears to measure what it purports to measure.
• Mean value the arithmetic average of a variable.
• Median value the value at the exact center when ranking cases of a variable.
• Measurement bias systematic over- or under-reporting of values for a variable.
• Mode the most frequently occurring value of a variable.
• Outlier a case with an extreme value for a variable relative to others.
• Reliability the consistency of results from the same measurement rules on the same case.
• Standard deviation a measure of the dispersion of a variable around its mean.
• Variance a measure of the dispersion of a variable around its mean.
• Variation the distribution of a variable's values across cases.
• Measurement validity the degree to which measurement captures the theoretical concept intended.
• External validity the extent to which findings can generalize to a broader population.
• Internal validity the accuracy of establishing a causal relationship between variables.
• Descriptive rq describes what something is like.
• Normative rq addresses how something should be.
• Explanatory rq explains why something is as it is, aiming at causality.
• Typical case average case where observed covariation exists; used for theory-testing.
• Deviant case case with unexpected Y value, used for theory-building or validation.
• interviewer effect interviewer’s personality and respondent perception influence responses.
• Surveys shorter, with closed-ended questions and larger sample sizes, analyzed quantitatively.
• Census targets an entire population.
• Representative surveys survey type where the sample that represents the target population of interest are chosen.
• Longitudinal/panel surveys the same respondents complete the survey multiple times over time.
• comparative surveys different cross-sectional units complete the same survey.
• Survey experiments survey that randomly assigns respondents to different treatments, comparing responses across groups.
• Sampling process of selecting units into a sample, representing a subset of the population.
• Units of analysis individuals, households, or communities in a study.
• Sampling frame list of all units in the population for sampling.
• Sampling strategies methods like simple random sampling, stratification, and cluster sampling are called...?.
• Sampling error random differences between sample and population.
• Sampling bias systematic difference between study participants and target population.
• Random error random factors affect responses, increasing data variation and threatening reliability.
• Systematic error factors that affect responses systematically, impacting measurement validity.
• Social desirability bias respondents answer in ways they believe are socially acceptable.
• Sensitivity bias respondents avoid truthful answers due to social stigma or sensitivity.
• Recall bias respondents struggle to remember events accurately under stress.
• Null hypothesis A theory-based statement about what we would observe if there were no relationship between an independent variable and the dependent variable
• Theory A tentative conjecture about the causes of some phenomenon of interest, idea or set of ideas that explains how or why something happens, based on evidence and observations
• Spurious Not what it appears to be, or false
• Parsimony Means that the theory explains more using less
• Observational study Involves watching and analyzing subjects without manipulating variables
• Population The entire group of cases you want to draw conclusions about
• Random sampling A method for selecting individual cases for a study in which every member of the underlying population has an equal probability of being selected
• Temporal order To infer that one variable causes another, researchers need to demonstrate that the cause precedes the effect in time
• Isolation refers to the process of controlling or minimizing the influence of confounding variables to accurately assess the relationship between specific variables of interes
• Mechanism Chain of events or reactions or changes, that connects the independent (cause) variable(s) with the dependent (outcome) variable
• Causal effect Requires covariation, temporal order, and isolation
• Causal criteria 1. Covariation 2. Temporal order 3. Isolation 4. Mechanism
• Inference The process of drawing conclusions or making judgments based on empirical evidence
• Intersubjectivity Scientific research is transparent, replicability is a related concept
• Control variable Variables that are held constant to prevent them from influencing the outcome
• Mediating variable A variable that explains the causal mechanism
• Discrete variables Are counted, can only take on certain values (number of fatalities in armed conflict)
• Triangulation At least two independent sources are confirming or supporting the same finding, combine different types of sources

Alla Inga Spara