Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
20 Cards in this Set
- Front
- Back
define:
parameter |
a number that describes some characteristic of a population
this value is only obtainable if all individuals in the population are measured for the variable of interest |
|
define:
sample statistic |
a number that describes the sample with regard to the variable of interest
|
|
define:
bias |
any systematic deviation of values of sample statistics away from the population parameter
|
|
what are two sources of random variation
|
1) measurement error
2) random sample variation |
|
what are three ways to minimize bias
|
1) selecting representative samples
2) measuring procedure that adequately quantifies the variable of interest 3) personnel are well trained |
|
2 ways to maximize precision
|
1) large number of subjects in data sample
2) use consistent methodology to collect data |
|
define:
anecdotal data |
collected haphazardly, individual cases which may be outliers
|
|
two types of random sampling
|
1) completely random
2) random systematic |
|
what is the purpose of stratified sampling
|
to give correct weight to the sample statistic computed from each strata of a heterogeneous population
|
|
sample study and experiment study are different. what is the goal of a sample study
|
to estimate the value of a population parameter
|
|
what is the goal of an experimental study
|
determine the cause-effect relationship between treatment variable and response variable
|
|
define:
treatment variable |
conditions imposed on the study subject
|
|
define:
response variable |
characteristic of the study subjects that is hypothesized to change as a result of the treatment
|
|
what are the 3 conditions necessary for completely random experimental design (20)
|
1) only one treatment variable (multiple levels of treatment are ok)
2) completely random assignment is possible 3) all extraneous factors can be controlled |
|
define:
randomized block experimental design |
comparisons to determine the treatment effect are made within blocks composed of different groups. these groups are defined by variables extraneous to the study which influence response to treatment
|
|
define:
factorial experimental design |
two or more treatment variables that are thought to interact with one another to influence a single response variable. measurements of the response variable are collected for all possible combinations of different treatment variables and levels
|
|
define:
natural experiment |
compare samples from two or more populations in their natural environment.
|
|
define:
a summary statistic |
a summarization of the raw data. examples are averages and such
|
|
example of a Distribution:
summary statistics for a set of 1000 values might include picking 20 values that actually occur in the set and 20 values for the number of times that each of the first 20 values occurs in the set. thus 40 values summarizes the 1000 values and is easier to assimilate than 1000! |
none
|
|
what are the 5 characteristics of a distribution
|
1) center: the middle of a range of data values where the most frequently occuring values are often found
2) spread: the ammount of variability in the values away from the center 3) shape: the number of peaks and whether the distribution is symmetric 4) gaps: segments within the range of data values where no values occur 5) outliers: data values that are very different from the other values |