Statistics and Sampling - Lecture Flashcards

A comprehensive set of flashcards covering sampling design, study components, data types, measurement, surveys, and common biases.

What questions should you ask about the 'Source' in a study?

Is the source reputable? Is the journal peer-reviewed? Who funded the study? Could funders have a vested interest in the outcome?

What does the 'Researchers' component refer to in the seven components?

The people actively collecting data; their behavior or expectations can influence responses.

What does 'Individuals or Objects Studied' require in terms of representative sampling?

A representative sample that replicates the population demographics; specify state/country, major, year, and how they were selected.

What does 'Measurements' involve in the seven components?

Some questions are hard to measure; some are ambiguous; definitions and measurement quality matter.

What does 'Setting' include in a study?

When, where, and how data were collected (season/time of day, public/private/lab, method like email/phone/face-to-face).

What are 'Extraneous Differences' and how are they addressed?

Differences between groups other than the treatment; random assignment helps control them in experiments.

What is meant by 'Magnitude' in the context of study results?

The size of observed differences; avoid vague language and report meaningful, clear estimates (e.g., risk reduction numbers).

What is 'Random Assignment' and why is it used?

Assigning participants to treatment or control groups by chance to protect against systematic differences between groups.

What is a 'Frame' in sampling?

A list or method used to define the population from which the sample is drawn.

What is a 'Census'?

Survey attempting to include every unit in the population.

Differentiate 'Population' vs 'Sample' vs 'Unit'.

Population is the entire group of interest; Sample is a subset from which data are collected; Unit is a single individual/object (e.g., person, household, school).

What is a 'Representative Sample'?

A sample that reflects the population's characteristics, allowing generalization to the population.

What is an 'Observational study' vs an 'Experiment'?

E study: observational, no manipulation. Experiment: manipulation of one or more explanatory variables to observe the effect on a response variable.

What is 'Data' in research terms?

Information gathered about subjects; can be numerical (measurements) or categorical (types).

What is a 'Sample Survey'?

An observational study where a sample is used to estimate characteristics of the whole population.

What is a 'Unit' in data collection?

A single individual or object on which data are collected (e.g., person, household, school).

What is a 'Placebo' in experiments?

A fake treatment with no active ingredients used to monitor the placebo effect.

What is a 'Control Group' in experiments?

Receives no treatment or a placebo; used for comparison with the treatment group.

What is a 'Treatment Group'?

Receives the actual treatment being tested.

What is a 'Lurking Variable'?

A variable not included in the study that can affect the results.

What is a 'Confounding Variable'?

An extraneous variable related to both the explanatory and response variables that may explain observed differences.

What is an 'Interacting Variable'?

A variable that changes the way the explanatory variable affects the response.

What is 'Measurement Validity'?

Whether a measurement accurately reflects the concept of interest without distortion.

What is 'Reliability' in measurement?

Consistency of a measurement; results are reproducible under the same conditions.

What is 'Bias (Measurement Context)'?

Systematic deviation from the true value due to the measurement context or method.

What is 'Natural Variability'?

Inherent fluctuations in values across individuals over time, even without measurement error or bias.

What is a 'Case-Control' study?

An observational study comparing those with a condition (cases) to those without (controls), often retrospective.

What is a 'Meta-Analysis'?

A study that combines results from multiple studies on the same topic to draw a comprehensive conclusion.

What is a 'Case Study'?

Detailed examination of a single individual or group; not generally generalizable.

What is a 'Double-Blind' study?

Neither participants nor researchers know who receives which treatment.

What is a 'Single-Blind' study?

Participants do not know which treatment they receive; researchers may know.

What is a 'Randomized Controlled Trial' (RCT)?

Participants are randomly assigned to treatment or control groups to test treatment effects.

What is a 'Matched Pairs Design'?

Subjects are paired and each pair receives different treatments.

What is a 'Simple Random Sample'?

Every unit has an equal chance of being selected; every possible sample of the same size is equally likely.

What is a 'Stratified Random Sample'?

Population is divided into strata; random samples are drawn from each stratum to improve representation.

What is a 'Cluster Sample'?

Population divided into clusters; some clusters are randomly selected; all individuals in chosen clusters are surveyed.

What is a 'Systematic Sample'?

Select every kth item from a list, starting at a random point.

What is a 'Volunteer Sample'?

Participants self-select into the study; typically biased and not representative.

What is a 'Convenience/Haphazard Sample'?

Non-random sample chosen for ease of access; prone to bias and not typically representative.

What is 'Random Digit Dialing'?

Randomly generating telephone numbers to reach a representative sample.

What is the 'Margin of Error' in sampling?

A measure of precision of an estimate due to sampling; e.g., ±3%; about 95% of samples would yield an interval containing the true population value.

What is a 'Frame out of date' error?

Non-sampling error where the frame no longer reflects the population, leading to bias.

Open-ended vs Closed-ended questions—what's the difference?

Open-ended allows free-form responses; closed-ended provides fixed choices and is easier to analyze.

What is a 'Pilot Study'?

A small preliminary version of a study to test procedures, questions, and measurement tools.

What is a 'Bias' in survey design?

Tendency for measurements to misrepresent the truth due to wording, sampling, or data collection methods.

What is the 'Desire to Please' bias?

Participants give answers they think researchers want to hear rather than their true beliefs.

What is 'Ordering of Questions' bias?

The sequence of questions can influence how participants respond.

What is 'Confidentiality' vs 'Anonymous' in surveys?

Confidentiality means researchers know identities but won’t share; Anonymous means researchers do not know identities, increasing honesty.

What is a 'Five-number Summary'?

Minimum, first quartile (Q1), median, third quartile (Q3), maximum.

What is a 'Boxplot' used for?

A graphical representation of the five-number summary and potential outliers.

What is the 'Mean' in data?

The average value; sum of measurements divided by the number of measurements.

What is the 'Median' in data?

The middle value when data are ordered; if even, the average of the two middle values.

What is the 'Mode' in data?

The most frequently occurring value.

What is the 'Range' in data?

Difference between the highest and lowest value.

What is 'Standard Deviation'?

A measure of the spread of data around the mean.

What is 'Variance'?

The square of the standard deviation; a measure of data dispersion.

What does 'Skewness' describe?

Asymmetry in the distribution of data.

What does 'Symmetry' mean in distributions?

A distribution where two halves are mirror images; mean ≈ median in symmetric distributions.

What is 'Left-skewed' (negatively skewed) data?

Longer left tail; mean < median.

What is 'Right-skewed' (positively skewed) data?

Longer right tail; mean > median.

What is a 'Likert scale'?

An ordinal scale (e.g., strongly disagree to strongly agree) used for attitudes or opinions.

What is a “Categorical” Variable?

A variable that represents distinct categories without a natural order, such as gender or marital status.

What are the two subsets of a Categorical Variable?

Nominal, Ordinal

What is a “nominal” variable? What is an example?

A type of categorical variable that labels distinct categories without any inherent order. None are better/worse than the other, just different

Examples: Student Numbers, Major, ethnicity, home state

What is a “ordinal” Variable? What is an example?

A type of categorical variable that represents categories with a meaningful order or ranking, where the distances between the ranks may not be uniform.
Examples: Education level, satisfaction rating, class rank, likert.

What is a “measurement” Variable?

A type of variable that quantifies with meaningful numbers or traits, such as temperature, height, weight, age, etc.

What are two types of Measurement variables?

Interval and Ratio

What is an “interval” variable? What is an example?

A type of measurement variable where the difference between values is meaningful, but there is no true zero point, such as temperature in Celsius or Fahrenheit. Also, differences are consistent (30 apart) but NOT in a ratio, no multiplication or division.

What is an “Ratio” variable? What is an example?

A type of measurement variable where both differences and ratios between values are meaningful, such as weight or height, # of somethings, which includes a true zero point.

What is “continuous” vs. “discrete” and which variable does this effect?

Continuous variables can take any value within a range, while discrete variables can only take specific, separate values. This distinction affects variables in measurement, where continuous could include height or weight, and discrete could include the number of individuals.

What is the definition of “statistics?

Science of collecting, describing, and interpreting data. Involving understanding how data is gathered, summarized, and used to draw conclusions.

What is “Quantitative Data”?

Data that represents counts or measurements.

What is a “dotplot”?

Graph w/ dots representing data values

What is a “stem-and-leaf” plot?

Short-hand way of writing down a list of numbers

What is a “mean”

The average of a set of numbers, calculated by dividing the sum of the values by the total number of values.

What is the “median”?

The middle value of a data set when arranged in ascending or descending order, or the average of the two middle values if there is an even number of observations.

What is the “mode”?

The value that appears most frequently in a data set.

What is the “range”?

The difference between the highest and lowest values in a data set, providing a measure of variability.

What is “standard deviation”?

A statistic that measures the dispersion or spread of a set of values around the mean, indicating how much individual data points differ from the average.

What is “variance”? How do you calculate it?

Variance quantifies the degree of spread in a data set, calculated by squaring the standard deviation.

What are “quartiles”?

Quartiles divide a data set into four equal parts, each containing 25% of the data points, helping to describe the distribution and spread.

What is the “Lower Quartile” and what symboles?

The Lower Quartile, also known as Q1 or QL, is the median of the lower half of a data set, separating the lowest 25% of the data from the rest.

What is the “upper quartile” and what symbols?

The Upper Quartile, also known as Q3 or QU, is the median of the upper half of a data set, separating the highest 25% of the data from the rest.

What is the “Five-number summary”?

The Five-number summary consists of the minimum, lower quartile (Q1), median, upper quartile (Q3), and maximum of a data set, providing a quick overview of its distribution.

What is “skewness”?

Description of how the values in a data set are spread out

Describe “Skewed to the left”?

Negatively skewed, smaller values more spread out than larger, mean is smaller than the median.

Describe “Skewed to the right?”

Positively skewed, larger values more spread out than smaller, mean is larger than the median.