mmw pre fi

Flashcards covering data management principles, study design, sampling methods, experimental design, and the Chi-Square test.

Data

Raw information or facts that become useful when organized in a meaningful way; can be qualitative or quantitative in nature.

Data Management

Concerned with looking after and processing data, including field data sheets, data correction, preparation for analysis, and documentation/archiving.

Importance of Data Management

Ensures high data quality for correct conclusions, allows future data use, enables result integration with other studies, improves processing efficiency and data meaningfulness.

Census

Systematically acquiring and recording information about all members of a given population.

Sample Survey

Selecting a subset within a population to gain knowledge about the entire population.

Experiment

Performed with controlled variables to study their effect on observed variables, requiring the possibility of replication.

Observation Study

Appropriate when there are no controlled variables and replication is impossible, often using a survey.

Characteristics of a Well-Designed Survey

Must be representative, incorporate a chance (e.g., random number generator), use neutral wording, and control possible errors and biases.

Sampling Frame

The subset of a population available for measurement from which the sample will be selected.

Nonprobability Sampling

Sampling method where some population elements have no chance of selection, or the probability of selection cannot be accurately determined.

Convenience Sampling

A nonprobability sample where customers in a supermarket are asked questions.

Quota Sampling

A nonprobability sample where judgment is used to select subjects based on specified proportions.

Probability Sampling

It is possible to determine which sampling units belong to which sample and the probability that each sample will be selected.

Simple Random Sampling (SRS)

All samples of a given size have an equal probability of being selected and selections are independent.

Systematic Sampling

Dividing the target population into equal-sized strata and then randomly selecting one element from each stratum.

Stratified Sampling

Organizing the frame by distinct categories (strata) and sampling each stratum as an independent sub-population.

Cluster Sampling

Cheaper method of sampling by selecting respondents from certain areas or time periods.

Matched Random Sampling

Two samples where members are paired or matched explicitly, or the same attribute is measured twice under different circumstances.

Characteristics of a Well-Designed Experiment

Includes stating the research purpose, designing experiments using blocking and randomized assignment, examining data for new hypotheses, and documenting results.

Control Groups and Experimental Units

Subjecting different groups to different conditions; an experimental unit is subjected to treatment, while a control group is not.

Random Assignments

Random allocation of (controlled variables) treatments to units.

Replication

Repeating measurements to reduce variability; the experiment itself should allow for replication to be checked by other researchers.

Confounding Variable

Extraneous variable that correlates with both dependent and independent variables, leading to potential false positive errors.

Placebo

Imitation pill identical to the actual treatment pill but without the treatment ingredients.

Placebo Effect

Sham effect whereby medical intervention has no direct health impact but results in improvement due to patient's knowledge of being treated.

Blinding

Technique used to make the subjects "blind" to which treatment is being given.

Blocking

Arranging experimental units into similar groups (blocks) to control for variability.

Completely Randomized Designs

Studying the effects of one primary factor without considering other nuisance variables; levels of the primary factor are randomly assigned to experimental units.

Randomized Block Design

Collection of completely randomized experiments, each run within one of the blocks of the total experiment.

Matched Pairs Design

Special case of randomized block design where blocks consist of just two elements (measurements on the same patient or similar patients).

Chi-Square Test

Used to determine if there is a significant difference between expected and observed frequencies in one or more categories.

Chi-Square Goodness of Fit Test

Determines if sample data matches a population.

Chi-Square Test for Independence

Compares two variables in a contingency table to see if they are related.