Back to the Table of Contents
An Introduction to Statistics
Review of Statistics Lessons 1 and 2
Terms & Vocab., Lesson 1: Data & Measure.
- Descriptive Statistics characterizes or describes a data set.
- Inferential Statistics tries to infer information about a population from a sample.
- Statistics is a collection of methods used in planning an experiment and analyzing data.
- Population is the complete set of data elements.
- A sample is a selected portion of a population.
- Parameters characterize a population, whereas a statistic is a sample measure.
- Accuracy is a measure of rightness, whereas precision is a measure of exactness.
- Statistics can be misused is a variety of ways to prove most anyone's point of view.
- Data are plural, whereas datum is singular.
- Qualitative data are nonnumeric: good, better, best.
- Quantitative data are numeric and can be either discrete (quantized) or continuous.
- Being able to do simple math: fractions, percentages, etc., is important.
- There are four levels of measurement: nominal, ordinal, interval, ratio.
- Ratio is the highest level, data are interval and has a starting point (zero, like Kelvin).
- Interval data have meaningful intervals between measurements (Celsius, Fahrenheit).
- Ordinal data have order but lack meaningful intervals: (strongly) agree, disagree, etc.
- Nominal data have names only: brown, plaid, paisley.
Terms & Vocabulary, Lesson 2: Sampling
- Sample size is very important.
- Measure, don't ask.
- Random errors are ok, systematic errors need to be accounted for,
sampling errors should be designed out. No randomization, no generalization.
- Sampling medium used (mail, phone, e-mail, personal interview) will affect accuracy.
- Sample must be representative of the population (avoid bias).
- Observational studies are more passive whereas experiments deliberately
impose treatments on individuals.
Can't always experiment. Experiments allow conclusions!
- There are five primary sampling methods.
- In random sampling any population member has a equal chance of being measured.
- In systematic sampling every kth member of the population is sampled.
- In stratified sampling the population is divided into two or more strata and each
subpopulation is sampled.
- In cluster sampling a population is divided into clusters and a few of
these clusters are exhaustively sampled.
- In convenience sampling the element can often select whether or not it is sampled.
- Be very wary of convenience sampling since it is prone to bias.
- Questions may be open ended (essay) or closed (multiple-choice, true/false).
- Studies may be retrospective (looking back) or prospective (looking ahead).