Terms  Definitions 

1/p 
Geometric mean

Bimodal 
Having two modes

Unimodal 
Having one mode

Parameter 
Measure of a population

Blocking 
Experimentation where groups of subjects are divided into "blocks" and tested upon within each of these blocks

Multimodal 
Having more than two modes

Random variable 
Represents the different numbers associated with potential outcomes of certain scenarios

Continuous 
Random variables that can assume values throughout the whole line interval

Histogram 
Graphical display where relative area corresponds to relative frequency

Quantitative displays 
Term which includes histograms, cumulative frequency plots (ogives), and stemplots

Categorical Variable 
All variables that name categories

Median (and IQR) 
Resistant to extreme values (outliers)

True 
Natural variation in outcomes can be reduced by using more subjects  True or False?

Experimental units 
Objects which an experiment is conducted upon; called "subjects" when they are human

Confounded 
When there is uncertainty with regard to which variable is causing an effect

Conditional probability 
Probability of an event happening given that another event has occurred

Doubleblinding 
When both the subjects nor those evaluating their responses know which treatment the subject has received

tdistribution 
We use this distribution because of the increase in the variability of the test statistic due to the estimation of the population standard deviation

Contingency Table 
Displays counts (%) of individuals falling into categories on two or more variables

Multistage sampling 
Type of sampling which is a combination of two or more types of sampling

[(s^2)/n + (s^2)/n]^0.5 
Drawing one standard deviation value from standard deviations of two independent samples

Relative frequency 
Number of times an occurrence happened divided by total number of trials

Placebo effect 
Psychological response to any kind of perceived treatment, even if treatment is simply a placebo

Clusters, Outliers 
Two major things to note when analyzing scatterplots

Deciles 
Values that lie up 1/10th and 9/10th of the way up a list; have percentile ranks of 10% and 90%

Discrete 
Random variables that can only be a counted number

StemandLeaf Display 
Shows quantitative data values in a way that sketches the distribution of the data

Skewed 
When a graph is not symmetric, you look at which tail stretches out further

Randomized paired comparison design 
Experimentation in which subjects are paired together and which subjects receiving which treatment is randomized

False 
A rvalue of 0 indicates there is no relationship between two variables  True or False?

Outliers on xaxis 
Which type of outliers are most influential on the LSRL?

Controlled random experiment 
What can indicate cause & effect relationships?

Frequency Table 
A table that lists categories in a categorical variable and gives counts of observations for each category

(a) increase alpha, (b) increase sample size, (c) decrease variation (st. deviation, variance, etc.) 
Three ways to increase power

df = n  2 
Degrees of freedom for Confidence Interval for population slope of LSRL

Quantitative Variable 
A variable in which the numbers act as numerical values

(a) 2 outcomes per event (b) independent observations (c) probability of success is constant (d) varying number of trials 
Factors of a geometric setting

Simple random sample (SRS) 
Way of selecting a sample so that each subject has equal chance of being chosen and that every possible of any size has an equal chance of being chosen

If the car speed was 0 MPH, we predict the mileage to be approximately 30 MPG. 
Interpret a yintercept of 30 in the relationship of speed vs. mileage

Simulation 
A pretend experiment

bias 
systematically favors certain outcomes

numerical 
each observation is a number

null hypothesis 
hypothesis to be tested

y hat 
variable representing the predicted value of the response variable

1st Quartile 
25% of data lies below

population 
(statistics) the entire aggregation of items from which samples can be drawn

Probability 
that chance of an event occurring

undercoverage 
some portion of the intended population is either unrepresented or underrepresented

Response Variable 
Measures an outcome of a study

replication 
ensuring that there is an adequate number of observations on each experimental treatment

Marginal Distribution 
In a contingency table, the distribution of either variable alone

placebo 
something presented as a drug (or whatever you are trying to test in your experiment), but having no actual effect on the person.

Sample 
The part of the population actually examined in order to gather information.

mean of a density curve 
The balance point.

factor 
a variable whose levels are controlled by the experimenter to discover its effects on the subjects

Observational Study 
observes individuals but does not attempt to influence the response. No treatment!!

correlated 
degree to which two variables are associated

Sampling 
Studying a part in order to gain information about the whole

Confounding 
When the levels of one factor are associated with the levels of another factor in such a way that their effects cannot be seperated.

Control 
A sample that is treated like other experimental groups except that the independent variable is not applied to it

stratified random sample 
first divide the population into groups of similar inviduals called strata

voluntary rsponse 
bias when people can chose to responde

independent events 
when the probability or occurence of one event has no effect on the other event

Distribution 
a display of the data WITH LABELS!!! .They can be different shapes...(skewed left, symmetric, skewed right)

standard deviation 
A measure of spread that describes an average distance from the mean.

Slope 
B1 gives a vlue in y units per x

linearity 
the notion that one event is a cause and the other is its effect; in behavior, the idea that one behavior is a stimulus and the other a response. Linear thinking is opposite of circular thinking, in which events are thought to be related in a series of interacting loops

Control group 
a group in an experimental study that is not given any specific treatment

dotplot 
a display for either kind of data that uses a dot to represent each individual in the data set

Simpsons Paradox 
: the reversal of some apparent trend when data are further explored or when groups are combined.

extraneous factor 
a variable that is not of interest in the current study but is thought to affect the response variable

Voluntary Response Bias 
When we receive biased results due to the responses received

prospective study 
a study where we collect data as it happens

planning and conducting a study 
1. understand the nature of the problem
2. decide what & how to measure 3. collect data 4. summarize & do preliminary analysis 5. do formal data analysis 6. interpret results 
Population Distribution of a Variable of Interest (or, more simply, The Population Distribution) 
The distribution of the variable of interest throughout a given population. Namely, each individual in a given population is measured or observed and the resulting distribution is simply called the population distribution.

Leave a Comment ({[ getComments().length ]})
Comments ({[ getComments().length ]})
{[ comment.comment ]}