Term
|
Definition
| The values or individuals of interest. Generally the whole population is not included in a study, due to sheer logistics. |
|
|
Term
|
Definition
| Subset of the population. The subjects of an experiment. Used because the population may be too large, or due to time or money restrictions it may be impractical to use the whole population. |
|
|
Term
|
Definition
| Numerical and graphical summaries of data used to convey results of a study. |
|
|
Term
|
Definition
| The possible values of the variable and the probability of the variable and the probability of each value occurring. Basically the spread of the data. |
|
|
Term
|
Definition
| Summaries of the distribution of the population. |
|
|
Term
|
Definition
| Summaries of the distribution of the sample. |
|
|
Term
|
Definition
| Information collected from the sample is generalized to the population. |
|
|
Term
|
Definition
| The objects described by a set of data. |
|
|
Term
|
Definition
| Any characteristic of an individual. |
|
|
Term
| Categorical (Qualitative)Variable |
|
Definition
| Places individuals into categories or groups. Examples: Eye Color, Race, or Type of Bike. |
|
|
Term
| Quantitative (Numerical) Variable |
|
Definition
| Takes numerical values for characteristics of an individual that would make sense for arithmetic operations. Example: Height, Distance, Number of gumballs chewed. |
|
|
Term
|
Definition
| Countable number of items. Example: Cities visited. |
|
|
Term
|
Definition
| Non-Discrete Quantitative variable. Example: Height, Temperature. Basically any measurement that is not a count is continuous. |
|
|
Term
|
Definition
| The act of process of investigating something. |
|
|
Term
|
Definition
| The population is a well defined, finite group of objects. |
|
|
Term
|
Definition
| Study in which the population is infinite or conceptual. |
|
|
Term
|
Definition
| Investigator's role in passive. They do not interfere in the outcome, and simply observe. Usually done when control is impossible or unethical. |
|
|
Term
|
Definition
| More common type of study. The investigator's role is active and they manipulate variables to study their effects. These studies are performed in order to establish causation. |
|
|
Term
|
Definition
| Name for individuals or subjects. |
|
|
Term
|
Definition
| Whatever is applied to the experimental units. |
|
|
Term
|
Definition
| The variable of interest. |
|
|
Term
|
Definition
| The relationship between two events. |
|
|
Term
|
Definition
| A variable that may have an important effect on the relationship in the variables in the study, but is not included among the variables studied. |
|
|
Term
|
Definition
| Randomly selecting the subjects in a study so that they are indicative of the population. |
|
|
Term
|
Definition
| List of all elements/ Units to be sampled. |
|
|
Term
|
Definition
| Individuals in the sample are self-selected. Also called voluntary response surveys. |
|
|
Term
|
Definition
| Average value. More sensitive to extreme values than median. |
|
|
Term
|
Definition
| Middle observation. Less sensitive to extreme values than mean. |
|
|
Term
|
Definition
|
|
Term
|
Definition
| Distributions in which the mean = median = mode. The histogram likely has a perfect bell shape. |
|
|
Term
| Positively or Right Skewed |
|
Definition
| Mean > Median. Right side of the histogram in stretched. |
|
|
Term
| Negatively or Left Skewed |
|
Definition
| Mean < Median. Left side of the histogram is stretched. |
|
|
Term
|
Definition
| S^2 = (1/(n-1))* Sum from i = 1 to n of (xi - xbar)^2 Strongly affected by outliers. |
|
|
Term
|
Definition
| S in the equation S^2 = (1/(n-1))* Sum from i = 1 to n of (xi - xbar)^2 Strongly affected by outliers. |
|
|
Term
|
Definition
| Q3-Q1, gives the spread of the middle 50% of the data. |
|
|
Term
|
Definition
| Most common graphical summary of quantitative data. Describes the observed distribution of a variable, also graphs the relative frequencies of a single quantitative variable. Uses intervals that covers the entire range of data. Draws bars to indicate the number of observations in each range. |
|
|
Term
|
Definition
| One peak in the histogram. |
|
|
Term
|
Definition
| Two peaks in the histogram. |
|
|
Term
|
Definition
| Gives less detail than a histogram. Sometimes called whisker plot. Shows the locations of min/max and Q1-Q3. |
|
|
Term
|
Definition
| The original hypothesis that motivates the experiment |
|
|
Term
|
Definition
| The process of collecting data with the aim of answering or exploring the conjecture. |
|
|
Term
|
Definition
| Statistical summary of the data from the experiment. |
|
|
Term
|
Definition
| What has been learned from the experiment. |
|
|
Term
|
Definition
| The variable that the investigator exercises control over. |
|
|
Term
|
Definition
| Supervised variable with one setting, or held constant. |
|
|
Term
|
Definition
| Supervised variable with several settings. |
|
|
Term
|
Definition
| Variable that is observed but is not a response variable or supervised variable. |
|
|
Term
|
Definition
| Categorical variables whose effect on the response variable is what we want to investigate. |
|
|
Term
|
Definition
|
|
Term
|
Definition
| The values of a factor in single factor experiment, or the combinations of levels of each factor in a multi-factor experiment. |
|
|
Term
|
Definition
| Variability among response values for experimental units that receive the same treatment. |
|
|
Term
|
Definition
| In experimental design this means that the experimental units are randomly allocated to the treatment groups. It makes large imbalances very unlikely. Protects the results from systematic influence from lurking variables. Does not prevent them from affecting the response, just the results. Does not reduce experimental error, but usually will average out over treatments. |
|
|
Term
| Completely Randomized Design (CRD) |
|
Definition
| Design in which all the experimental units are randomly assigned to treatments. Every unit has the same chance of receiving any treatment. |
|
|
Term
|
Definition
| Many experimental units that receive the same treatment. Helps generalize results to the population. Quantifies the amount of experimental error. |
|
|
Term
|
Definition
| Homogeneous blocks of units for which nuisance factors are held constant, and the factor of interest is allowed to vary. Prevents the nuisance variable from affecting the results, and reduces the experimental error within groups. |
|
|
Term
|
Definition
| The fitted main effect for factor A at its ith level is ai = ybari - ybar |
|
|
Term
|
Definition
| Same number of replicates for each treatment. |
|
|
Term
|
Definition
| (ab)ij = yij - (ybar + ai + bj) Measures how the factors interact. If all (ab)ij's are close to zero then this indicates that there are no interaction effects. |
|
|
