Glossary of Key Terms
Understanding the language of survey research and data analysis is crucial for making informed decisions. Our glossary covers essential terminology used in our jury analysis process.
- Confidence intervals: A range of values, derived from sample data, that is likely to contain the true population parameter. A confidence interval represents the margin of error and can often be depicted on a graph to communicate uncertainty effectively.
- Double blind: A method of assigning experiment participants to experimental conditions where neither the participant nor the research know what condition the participant is assigned to. Double blind assignments address the possibility that participants will behave differently knowing their role in an experiment.
- Hypothesis testing: A statistical method used to determine whether there is enough evidence in a sample to infer a particular outcome for the population.
- Jury-qualified adults: Adults who meet qualifications for serving on juries in the United States, such as U.S. citizenship, no felony convictions, and physical and mental ability to serve on jury. There are additional qualifications in death penalty cases.
- Margin of error: An expression of the range within which we expect the true population parameter to fall, based on the survey results.
- Randomized experiment: A research method where the research randomly assigns participants to experimental conditions with varying levels of some causal variable hypothesized to affect measurable outcomes. Randomized experiments are considered the gold standard of proof because the design ensures that the only difference between experimental groups is the causal variable and nothing else explains observed differences in measured outcomes.
- Reliability: The consistency of a measurement or survey tool. A reliable survey will produce the same results under consistent conditions. In jury analysis, reliability ensures that your methods yield stable, repeatable results over time.
- Sampling frame: The population from which a sample is drawn. It represents the population you wish to study. In our research, the sampling frame excludes adults who do not meet basic jury qualifications, even though they may be part of the sampling frame for other research purposes.
- Sample size: The number of individuals surveyed to draw conclusions about the larger population.
- Statistic vs. Parameter: A statistic is a numerical value calculated from a sample, such as the mean or proportion. A parameter, on the other hand, is a numerical value that describes a characteristic of the entire population. While a statistic is used to estimate a parameter, the true parameter is often unknown.
- Statistical power: The probability that a test will detect a true effect or difference when it exists. High statistical power means there’s a lower chance of missing a real effect, which helps ensure that your analysis produces reliable results. Power is affected by factors such as sample size, effect size, and significance level.
- Statistical significance: The determination whether the results of a study could have occurred by random chance, as opposed to a genuine effect or difference. A result is statistically significant if it passes a pre-determined threshold (usually 0.05), suggesting that the observed effect is likely genuine.
- Validity: The extent to which a measurement is an unbiased measure of what it is intended to measure. In the context of jury analysis, a valid survey accurately reflects jurors’ attitudes or likely behaviors in relation to a case.