D Glossary

\(68\)--\(95\)--\(99.7\) rule: For any bell-shaped distribution, approximately \(68\)% of observations lie within one standard deviation of the mean, \(95\)% of observations lie within two standard deviations of the mean, and \(99.7\)% of observations lie within three standard deviations of the mean. Also called the empirical rule.
Accuracy: Accuracy refers to how close a sample estimate is likely to be to the population value, on average.
Alternative hypothesis: The alternative hypothesis proposes that any difference, change or relationship observed in the sample is because a difference, change or relationship exists the population (that is, the difference, change or relationship cannot be explained by sampling variation).
Bell-shaped distributions: See Normal distributions.
Between-individual comparisons: See Comparison (between individuals), Comparison.
Bias: Bias refers to anything systematic which may cause differences between samples statistics and population parameters.
Blinding: Blinding when those involved in the study do not know which comparison group the study individuals are in.
A study can blind the researcher to knowing what comparison group the study individuals are in.
A study can blind the participants to knowing what comparison group they are in.
A study can blind the analysts to knowing what comparison group the individuals are in during analysis.
Blocking: Blocking occurs when units of analysis are arranged in separate groups of similar units (called blocks).
Carry-over effect: The carry-over effect occurs when the influence of one treatment or conditions influences individuals' responses to subsequent treatments or conditions. (Sect. 7.7) or observational studies (Sect. 8.5).
Cases: The individuals units in the population; the units of analysis. Also called individuals, or subjects when the individuals are people.
Categorical data: See qualitative data.
Chi-square (\(\chi^2\)) score: The test-statistic used when comparing odds (or proportions). The \(\chi^2\) statistic measures the overall size of the differences between the expected counts and observed counts, over the entire \(2\times 2\) table.
Classical approach to probability: In the classical approach to probability, the probability of an event occurring is the number of elements of the sample space included in the event, divided by the total number of elements in the sample space, when all outcomes are equally likely.
Cluster sampling: A sample where the population is split into a large number of small groups called clusters, then a simple random sample of clusters is selected and every member of the chosen small groups is part of the sample.
Collusion: Collusion occurs when people work together to produce work, but the work is unfairly presented to give an advantage to some of the group.
Comparison: In a RQ, a comparison refers to groups of individuals being compared. The comparison may be within individuals, or between groups of individuals. Also see Comparison (between individuals), Comparison (within individuals).
Comparison (between individuals): The between-individuals comparison in the RQ identifies the small number of different groups of individuals for which the outcome is compared. Also see Comparison (within individuals), Comparison.

Comparison (within individuals) Also see Comparison (between individuals), Comparison.

Compound event: A compound event is any combination of simple events (i.e., of elements in the sample space).
Conceptual definition: A conceptual definition articulates precisely what words or phrases mean in a study.
Conditions: The conditions are the values of the comparison that those in the observational study experience, but are not imposed by the researchers.
Confidence interval: A confidence interval is an interval in which the population parameter is likely to be contained, if we found many samples the same way.
If we computed the \(95\)% confidence interval (or CI) from each sample, about \(95\)% of the CIs would contain the statistic of interest. This interval is called a confidence interval.
Alternatively, the CI is the range of plausible values for the parameter that may have produced the observed sample statistic. We studied CIs is some specific situations (there are hundreds more!):

CIs for one proportion: Chap. 24
CIs for one mean: Chap. 25
CIs for a mean difference (paired sample mean): Chap. 27
CIs for the difference between two means: Chap. 28
CIs for comparing two odds: Chap. 29
CIs for regression parameters: Sect. 39.6

Confounding: Confounding is when a third variable influences the observed relationship between the response and explanatory variable.
Confounding variable: A confounding variable (or a confounder) is an extraneous variable associated with the response and explanatory variables.
Conditions: The conditions are the values of the explanatory variable that those in the observational study experience, but are not imposed by the researchers.
Continuous data: Continuous quantitative data has (at least in theory) an infinite number of possible values between any two given values.
Control: A control is a unit of analysis without the treatment or condition of interest, but as similar as possible in every other way to other units of analysis.
Convenience sample: A sample where individuals are selected because they are convenient for the researcher.
Data: Data refers items of information obtained from a study (such as height of seedlings, or the type of medication given).
Dataset: A dataset refers to a collection of data.
Descriptive research question: Descriptive RQs have population and an outcome.
Descriptive study: Descriptive studies answer descriptive research questions.
Discrete data: Discrete quantitative data has a countable number of possible values between any two given values of the variable.
Distribution: The distribution of a variable describes what values are present in the data, and how often those values appear.
Ecological validity: A study is ecologically valid if the study methods, materials and closely context approximate the real situation of interest.
Event: An event is any combination of the elements in the sample space.
Exclusion criteria: Exclusion criteria are characteristics that disqualify potential individuals from being included in the study.
Empirical rule: For any bell-shaped distribution, approximately \(68\)% of observations lie within one standard deviation of the mean, \(95\)% of observations lie within two standard deviations of the mean, and \(99.7\)% of observations lie within three standard deviations of the mean. Also called the \(68\)--\(95\)--\(99.7\) rule.
Experimental studies (or Experiments): Experimental studies (or experiments) study relationships with an intervention.
Experimenter effect: See Observer effect
Explanatory variable: An explanatory variable may (partially) explain or be associated with changes in another variable of interest (the response variable). In an experimental study, it is the variable that can be manipulated by the researchers.
External validity: External validity refers to the ability to generalise the results of the study to the rest of the population, beyond just those in the studied sample. For a study to be truly externally valid, the sample must be a random sample from the population.
Extraneous variable: An extraneous variable is any variable associated with the response variable, but is not one the explanatory variable.
Extrapolation: Extrapolation refers to making prediction outside the range of the available data. Extrapolation beyond the data may lead to nonsense predictions.
Fraud: Fraud refers to the intent to deceive. Fraud can occur by:

taking an exam for another student or letting someone take an exam for you
falsifying or inventing research data and findings
altering or fabricating information
forging a document
falsifying past academic records or employment details in order to gain entrance into the university

Hawthorne effect: The Hawthorne effect is the tendency of individuals to change their behaviour if they know (or think) they are being observed, in experimental studies (Sect. 7.4) or observational studies (Sect. 8.3).
Hypothesis: A hypothesis is a possible answer to a (research) question. More specifically, see null hypothesis or alternative hypothesis
Hypothesis test: A hypothesis test is a way to formally answer questions about a population, based on information obtained from a sample. In this book, we have looked at some specific hypothesis tests:

Hypothesis tests about a single mean: Chap. 32
Hypothesis tests about a mean difference (means of paired samples): Chap. 34
Hypothesis tests comparing two means: Chap. 35
Hypothesis tests comparing odds (or percentages): Chap. 36
Hypothesis tests about a correlation: Sect. 38.2
Hypothesis tests about regression parameters: Sect. 39.7

Inclusion criteria: Inclusion criteria are characteristics that individuals must meet explicitly to be included in the study.
Independence: Two events are independent if the probability of one event doesn't change depending on whether or not other event has happened.
Individuals: The individuals units in the population from which the observations of interest could be made; the units of analysis. Also called cases, or subjects when the individuals are people.
Internal validity: Internally validity refers to the extent to which a cause-and-effect relationship can be established in a study.
A study with high internal validity shows that the changes in the response variable can be attributed to changes in the explanatory variables; other explanations have been ruled out.
Intervention: An intervention is present when researchers can manipulate (or impose) the explanatory variable on the individuals to determine the impact on the response variable.
IQR: The IQR is the range in which the middle \(50\) of the data lie; the difference between the third and the first quartiles.
IQR rule for identifying outliers: The IQR rule can identify outliers as either:

mild (observations \(1.5\times \text{IQR}\) more unusual than \(Q_1\) or \(Q_3\)), or
extreme (observations \(3\times\text{IQR}\) more unusual than \(Q_1\) or \(Q_3\)).

Jittering: Jittering is when a small amount of randomness is added in either the horizontal or vertical direction (or sometimes both) to separate points that would otherwise be over-plotted.,
Judgement sample: A sample where individuals are selected, based on the researchers' judgement, depending on whether the researcher thinks they are likely to be agreeable or helpful.
Levels of a qualitative variable: The levels (or the values) of a qualitative variable refer to the names of the distinct categories.
Lurking variable: A lurking variable is an extraneous variable associated with the response and explanatory variables (that is, is a confounding variable), but whose values are not recorded in the study data.
Mean: The mean is one way to measure the 'average' value of quantitative data. The arithmetic mean is the 'balance point' of the data, and the value such that the positive and negative distances from the mean add to zero.
Median: The median is one way to measure the 'average' value of some data. The median is a value such that half the values are larger than the median, and half the values are smaller than the median.
Mode: A mode is one way to measure the 'average' value of some qualitative data. A mode is the level (or levels) of a qualitative variable with the most observations.
Multistage sampling: A sample where large groups are selected using a simple random sample, then smaller groups within those large groups are selected using a simple random sample. The simple randomly sampling can continue for as many levels as necessary.
Nominal variable: A nominal qualitative variable is a qualitative variable where the levels do not have a natural order.
Normal distribution: A normal distribution is symmetrical distribution, with most values in the centre of the distribution. The normal distribution is described by its mean and standard deviation. A picture of a normal distribution is shown in Fig. D.1. Normal distributions are also called bell-shaped distributions.

FIGURE D.1: A normal distribution

Null hypothesis: The null hypothesis proposes that any difference, change or relationship observed in the sample can be explained by sampling variation (that is, no difference, change or relationship exists the population).
Observational study: Observational studies study relationships without an intervention.
Observer effect: The observer effect occurs when the researchers (unconsciously) change their behaviour to conform to expectations because they know what values of the explanatory variable apply to the individuals. This may then cause the individuals to change their behaviour or reporting also.
Odds: The odds of some event is the proportion (or percentage, or number) of times that an event happens, divided by the proportion (or percentage, or number) of times that the event does not happen.
Odds ratio: The odds ratio is how many times greater the odds of an event are in one group, compared to the odds of the same event in a different group.
Operational definition: An operational definition articulates exactly how something will be identified, measured, observed or assessed.
Ordinal variable: An ordinal qualitative variable is a qualitative variable where the levels do have a natural order.
Outcome: The outcome in a RQ is the result, output, consequence or effect of interest in a study, numerically summarised for a group.
Outliers: An outlier is an observation that is 'unusual' (either larger or smaller) compared to the bulk of the data. Rules for identifying outliers are arbitrary.
\(P\)-value: A \(P\)-value is the probability of observing the sample results (or something even more extreme) over repeated sampling, under the assumption that the null hypothesis about the population is true.
Parameter: A parameter is a number, usually unknown, describing some feature of a population, and estimated by a statistic.
Paired data: Paired data occurs when every observation in one group is related to, or can be matched sensibly to, one unique observation in another group.
Percentage: A percentage is a proportion, multiplied by \(100\). In this context, percentages are numbers between \(0\)% and \(100\)%.
Percentiles: The \(p\)th percentile of the data is a value separating the smallest \(p\) of the data from the rest.
Pilot study: A pilot study is a small test run of the study protocol used to check that the protocol is appropriate and practical, and to identify (and hence fix) possible problems with the research design or protocol.
Placebo: A placebo is a treatment with no intended effect or active ingredient, but appears to be the real treatment.
Placebo effect: The placebo effect occurs when individuals report perceived or actual effects despite not receiving an active treatment or condition, in experimental studies (Sect. 7.6).
Plagiarism: Plagiarism is using other people’s ideas and research to develop new conclusions, or confirm existing conclusion. All sources used when writing research should be acknowledged, otherwise you are committing plagiarism.
Plagiarism can be deliberate or accidental.
Population: A population is a group of individuals (or cases, or subjects if the individuals are people) from which the total set of observations of interest could be made, and to which the results will (hopefully) generalise.
Precision: Precision refers to how similar the sample estimates from different samples are likely to be to each other (that is, how much variation is likely in the sample estimates).
Probability: A probability is a number between \(0\) and \(1\) inclusive (or between \(0\)% and \(100\)% inclusive) that quantifies the likelihood that a certain event will occur. A probability of zero (or \(0\)%) means the event is 'impossible' (will never occur), and a probability of one (or \(100\)%) means that the event is certain to happen (will always occur). Most events have a probability between the extremes of \(0\)% and \(100\)%.
Proportion: A proportion is a fraction out of a total. Proportions are numbers between \(0\) and \(1\).
Protocol: A protocol is a predefined procedure detailing the design and implementation of studies, and for data collection.
Qualitative data: Qualitative data is not mathematically numerical data: it consists of categories or labels (even if those labels are numbers). Also called categorical data.
Quantitative data: Quantitative data is mathematically numerical data: the numbers have numerical meaning, and represent quantities or amounts. Quantitative data generally arise from counting or measuring.
Quantitative research: Quantitative research summarises and analyses data using numerical methods, such as producing averages and percentages.
Quartiles: Quartiles describe the variation and shape of data:

The first quartile \(Q_1\): A value that separates the smallest \(25\)% of observations from the largest \(75\)%. The \(Q_1\) is like the median of the smaller half of the data, halfway between the minimum value and the median.
The second quartile \(Q_2\): A value that separates the smallest \(50\)% of observations from the largest \(50\)%. (This is the median.)
The third quartile \(Q_3\): The value that separates the smallest \(75\)% of observations from the largest \(25\)%.
The \(Q_3\) is like the median of the larger half of the data, halfway between the median and the maximum value.

Quasi-experiment: In a quasi-experiment, the researchers (1) allocate treatments to groups of individuals (i.e., do not allocate the values of the explanatory variable for the individuals), but (2) do not determine who or what is in those groups.
Questionnaire: A questionnaire is a set of questions for respondents to answer.
Random: In research and statistics, random means 'determined completely by impersonal chance'.
Random procedure: A random procedure is a sequence of well-defined steps that (a) can be repeated, in theory, indefinitely under essentially identical conditions; (b) has well-defined results; and (c) has result that are unpredictable from any individual repetition.
Range: The range is the maximum value of a variable minus the minimum value of the variable.
Relational research question: Relational RQs have population, outcome, and a between-individuals comparison.
Relative frequency approach to probability: In the relative frequency approach to probability, the probability of an event is approximately the number of times the outcomes of interest has appeared in the past, divided by the number of 'attempts' in the past. This produces an approximate probability.
Repeated-measures research question: Repeated-measures RQs have a population, outcome and a within-individuals comparison.
Representative samples: A representative sample is one where the individuals in the sample are not likely to be different the individuals not in the sample, at least for the variables of interest.
Research design: Research design refers to the decisions made by the researchers to maximise external validity and internal validity.
Research question: ????????
Response variable: A response variable records the result, output, consequence or effect of interest from changes in another variable (the explanatory variable).
Sample: A sample is a subset of individuals from the population from which data are collected.
Sample space: The sample space is a list of all possible and distinct results after administering a random procedure once.
Sampling distribution: A sampling distribution is the distribution of some sample statistic, showing how its value varies from one sample to sample.
Sampling frame: The sampling frame is a list of all the individuals in the population.
Sampling mean: The sampling mean is the mean of the sampling distribution of a statistic.
Sampling variation: Sampling variation refers to how the sample estimates (statistics) vary from sample to sample, because each sample is different.
Selection bias: Selection bias is the tendency of a sample to over- or under-estimate a population quantity.
Scale data: See quantitative data.
Simple event: A simple event is a single element of the sample space.
Simple random sample: A sample where every possible sample of a given size has same chance of being selected.
Standard deviation: The standard deviation is, approximately, the average distance of observations from the mean.
Standard deviation rule for identifying outliers: For approximately symmetric distributions, any observation more than three standard deviations from the mean can be considered an outlier.
Standard error: A standard error is the standard deviation of all possible values of the sample estimate (from samples of a certain size). Any quantity estimated from a sample has a standard error.
Stratified sampling: A sample where the population is split into a small number of large (usually homogeneous) groups called strata, then cases are selected using a simple random sample from each stratum.
Statistic: A statistic is a number describing some feature of a sample (to estimate the unknown value of the population parameter).
Statistical validity: A result is statistically valid if the conditions for the underlying mathematical calculations and assumptions to be approximately correct are met. Every confidence interval and hypothesis test has statistical validity conditions.
Subjective approach to probability: In the subjective approach to probability, various factors are incorporated, perhaps subjectively, to determine the probability of an event.
Subjects: The individuals units in the population when they are people; the units of analysis. Also called individuals or cases; however, those two terms do not refer to people.
Systematic sampling: A sample where the first case is randomly selected; then, every \(n\)th individual is selected.
\(t\)-score: A \(t\)-score measures how many standard deviations a value is from the mean. A \(t\)-score is like a \(z\)-score, but uses sample information to estimate the standard deviation of the variable. Also see: \(z\)-score.
Treatments: The treatments are the values of the explanatory variable that the researchers impose upon the individuals in the experimental study.
True experiment: In a true experiment, the researchers (1) allocate treatments to groups of individuals (i.e., explanatory variable to the individuals), and and (2) determine who or what is in those groups. While the steps may not happen explicit, they can happen conceptually.
Unit of observation: The unit of observation is the 'who' or 'what' that are observed, from which measurements are taken and data collected.
Unit of analysis: The smallest collection of units of observations (and perhaps the units of observations themselves) about which conclusions are made; the smallest independent elements of the population for which information is analysed. In an experimental study, the unit of analysis is the smallest collection of units of observations that can be randomly allocated to separate treatments.
Unstandardizing formula: When the \(z\)-score is known, the unstandardising formula determines the corresponding value of the observation \(x\).
Values of a qualitative variable: See levels.
Variable: A variable is a single aspect or characteristic associated with the individuals, whose values can vary from individual to individual.
Voluntary response (self-selecting) sample: A sample where individuals participate if they wish to.
Within-individual comparison: See Comparison (within individuals),Comparison.
\(z\)-score: A \(z\)-score measures how many standard deviations a value is from the mean. In symbols:
\[ z = \frac{x - \mu}{\sigma}, \] where \(x\) is the value, \(\mu\) is the mean of the distribution, and \(\sigma\) is the standard deviation of the distribution. Also see: \(t\)-score.

C Symbols, formulas, statistics and parameters

E Answers to end-of-chapter exercises