Community Exam 3 Review

Biostatistics

The application of statistical methods to analyze public health problems and to further biomedical research.

Parameter

A numerical measure that describes a characteristic of an entire population, considered a true value.

Statistic

A numerical measure that describes a characteristic of a sample, serving as an estimate of the true population parameter.

Probability Sampling

Sampling procedures that incorporate a chance component where every participant has a known chance of selection, ensuring the sample reasonably represents the population.

Simple Random Sampling

The purest form of probability sampling, where each member of the population has an equal chance of being selected; considered the gold standard for minimizing sampling error.

Systematic Sampling

A probability sampling technique, also called an nth selection technique, in which every nth individual is selected from a population list.

Stratified Sampling

A probability sampling method in which the population is divided into homogeneous subpopulations (strata) and random samples are drawn from each stratum.

Cluster Sampling

the population is divided into heterogeneous clusters, clusters are randomly selected, and all subjects within the selected clusters are surveyed.

Nonprobability Sampling

Nonrandom sampling strategies used when probability; results cannot be generalized to the entire population.

Convenience Sampling

subjects are chosen based on their relative ease of access to the researcher, such as classmates or coworkers.

Quota Sampling

a population is first segmented into subgroups, after which subjects are selected based on a non-random, specific requirement.

Discrete Data

Quantitative data for which only a finite or countable number of distinct values (such as whole numbers) are possible.

Continuous Data

Quantitative data that can assume any numerical value along a continuous scale within a given range.

Variable

value or characteristic that can differ from individual to individual.

Nominal Level of Measurement

The lowest level of measurement, used strictly for labeling variables without quantitative value. (Names or Labels)

Ordinal Level of Measurement

A level of measurement consisting of categories that have rank order, but lack equal or defined distances between consecutive ranks.

Interval Level of Measurement

A level of measurement that orders measurements with equivalent distances between scale intervals, but possesses no true zero point (e.g., Celsius temperature).

Ratio Level of Measurement

The most precise level of measurement, featuring equal intervals and an absolute true zero where zero indicates the complete absence of the property. (Height)

Descriptive Statistics

The phase of data analysis focused on collecting, summarizing, organizing, and presenting data.

Inferential Statistics

The phase of data analysis focused on analyzing sample data to draw conclusions and generalize findings back to the broader population.

Median

best choice for an average if the distribution of values has extreme scores.

Positively Skewed Distribution

An asymmetrical distribution with an elongated tail extending to the right, where the mean is greater than (to the right of) the median and mode.

Negatively Skewed Distribution

An asymmetrical distribution with an elongated tail extending to the left, where the mean is less than (to the left of) the median and mode.

Measures of Dispersion (Measures of Variability)

are numbers that indicate how spread out a set of data is along a distribution

Range

the difference between the largest value and the smallest value in a distribution.

Variance

A measure of dispersion describing the average squared deviation of individual scores from the mean.

Standard Deviation

A measure of dispersion calculated as the square root of the variance that describes the spread or broadness of values around the mean.

Empirical Rule (Normal Distribution)

The statistical rule stating that approximately $68\%$ of values lie within $\pm 1$ standard deviation of the mean, $95\%$ within $\pm 2$ standard deviations, and $99.7\%$ within $\pm 3$ standard deviations.

Correlation

a relationship or association between variables that can be measured mathematically.

Correlation Coefficient ($r$)

A statistic ranging between $-1.00$ and $+1.00$ that quantifies the strength (numerical magnitude) and direction (positive or negative sign) of a relationship between two variables.

Linear Regression

A statistical procedure for estimating relationships among variables by plotting intersecting data points on a scatter diagram and fitting a straight line.

Causality

A relationship between an initial event (the cause) and a second event (the effect), where the second event is understood as a direct consequence of the first.

Primary Literature

Original reports of new information and discoveries that reflect original thinking, such as a primary research report or clinical trial.

Secondary Literature

Publications that interpret, evaluate, or review evidence previously reported in primary literature, such as most science textbooks.

Tertiary Literature

Publications that summarize and condense primary and secondary literature, including encyclopedias, dictionaries, manuals, and abstracts.

Hypothesis (Alternative Hypothesis Ha)

educated guess of a possible explanation to the research question.

Independent Variable

The variable manipulated or controlled by the researcher to determine that brings about a change in the dependent variable.

Dependent Variable

those changed or affected by the independent variable

Extraneous Variable (Confounding variable)

An independent variable not related to the study's purpose that may influence the dependent variable and confound the observed relationship.

Null Hypothesis ($H_0$)

A statement proposing that no relationship or significant difference exists between the independent and dependent variables, which the researcher aims to test and disprove.

Type I Error (Alpha Error)

The error that occurs when a researcher rejects the null hypothesis when it is actually true (a false positive).

Type II Error (Beta Error)

The error that occurs when a researcher fails to reject the null hypothesis when it is actually false (a false negative).

Acceptable level of making an alpha error

0.05 or 0.01 (medical research at 0.001)

Confidence Interval

A range of values around a sample statistic with a specified probability ($95\%$ when p=0.05) of containing the true population parameter.

Statistical Power ($P$)

The probability of correctly rejecting a false null hypothesis with an accepted conventional standard of $0.80$ ($80\%$).

Parametric Statistical Tests

Inferential statistical tests designed for quantitative (interval and ratio) data that assume scores follow a normal bell-shaped distribution.

Nonparametric Statistical Tests

Inferential statistical tests designed for qualitative (nominal and ordinal) data or when values are skewed

Student's t-Test

The most common parametric statistical test, used to compare the mean scores of two groups.

Dependent t-Test

A parametric test used to compare two mean scores derived from the same group of subjects at different times (e.g., pretest and posttest) or from paired subjects.

Independent t-Test

A parametric test used to determine whether there is a statistically significant difference between the means of two unrelated, independent groups.

Analysis of Variance (ANOVA)

A parametric statistical test used to test and compare the differences among three or more group means.

One-Way ANOVA

evaluating the effect of one independent variable across three or more groups.

Two-Way ANOVA

evaluating the simultaneous effects of two independent variables across groups.

Chi-Square Test ($\chi^2$)

A nonparametric statistical test used to evaluate relationships and associations between categorical variables arranged in rows and columns of a contingency table without comparing means.

Pretest-Posttest Design

An experimental study design where the dependent variable is measured prior to an intervention (baseline) and again following the intervention for comparison.

Repeated Measures Design

An experimental study design, also known as a time series design, in which the dependent variable is measured at multiple posttest intervals to evaluate whether treatment effects persist over time.

Crossover Design

An experimental design in which all participants receive each treatment in a different sequence, with groups switching treatments after an intermediate washout period.

Washout Period

A designated timeframe in a crossover experimental design during which no treatment is administered to eliminate carryover effects from the prior intervention.

Split-Plot Design

An oral health experimental design, also called a split-mouth design, where two or more treatments are applied to separate equivalent units (e.g., contralateral quadrants) within the same subject's mouth.

Factorial Design

An experimental study design that evaluates the simultaneous effects and interactions of two or more independent variables (factors) at multiple levels within the same study.

Single-Blind Design

A study design where research subjects are unaware of their assignment to the experimental or control group, but the experimenters know group assignments.

Double-Blind Design

A study design where neither the subjects nor the research personnel directly interacting with them know which participants receive the active treatment or the placebo.

Hawthorne Effect

A phenomenon where study subjects alter their behavior and perform better simply because they are aware they are participating in an experimental study.

Institutional Review Board (IRB)

An institutional committee of at least 5 members from diverse backgrounds formally designated to approve, monitor, and review human subject research to protect subjects' rights and welfare.

ad hoc committee

formed for a particular purpose

Frequency Tables

show frequency that values occur within a particular group.

Bar Graph

A graphic display using separated, non-touching bars of equal width to illustrate frequencies of nominal or ordinal data. (discrete)

Histogram

A graphic display of contiguous (touching) vertical bars used to depict the frequency distribution of continuous variables (interval or ratio)

Frequency Polygon

A continuous line graph used to portray continuous frequency data, allowing multiple distributions to be compared on the same axes.

Time Series Graph

A graph (most often a line graph) used to plot and illustrate changes in a measured variable across successive time points.

Pie Chart

A circular diagram divided into proportional sections illustrating numerical percentages of a whole; not recommended for scientific literature.

Journal Sections

Abstract, Introduction, Materials and Methods, Results, Discussion, Summary and Conclusions, References

Observational Study

An epidemiologic study design in which the researcher studies, records, and measures events without actively manipulating or altering the circumstances.

Descriptive Study

An observational study design that characterizes a health event by defining person (who), place (where), and time (when).

Analytical Study

An observational study design that compares groups to investigate causes, risk factors, and the reasons how and why a health event occurred.

3 Observational Studies

cross-sectional, cohort, and case-control

Cross-Sectional Study (Prevalence Study)

An observational study that assesses exposure and disease prevalence simultaneously in a defined population at a single point in time, providing a snapshot.

Case-Control Study

A retrospective observational study that compares individuals with a disease (cases) to individuals without the disease (controls) to examine historical exposure levels. (always retrospective)

Prospective Cohort Study

An observational study that follows disease-free cohorts classified by exposure status forward in time to observe the incidence of disease.

Retrospective Cohort Study

An observational study where cohorts are defined according to past exposure status and reviewed backward in time using existing records to assess disease occurrence.

Systematic Review

A structured, comprehensive synthesis of all primary research meeting prespecified eligibility criteria (“golf standard”)

Meta-Analysis

A statistical pooling technique that mathematically combines quantitative results from multiple independent studies within a systematic review to yield a single summary conclusion.