section 7-B data Flashcards
1
Q
data
A
- medium with which the behavior analyst works
- results of measurement
- empirical basis for decision making
2
Q
graphs
A
- visual format for displaying data
- reveal relations between measurements & variables
- how BA organize, store, interpret, and communicate the results of our work
3
Q
3 purpose of graphs
A
- communicate our data
- assess data
- show how DV/IV related
4
Q
5 benefits of graphs
A
- immediate picture of behavior
- allow explore interesting variations in B as they occurring
- judgment aid to help interpret results
- conservative method for determining the sig. of B change a B change that is statistically sig. may not look impressive on graph
- allow independent judgment & evaluation of data
5
Q
5 types graphs
A
equal-interval graphs:
- line
- bar
- cumulative record
- scatter plot
non-equal interval graphs:
- standard celebration chart (logarithmic scales)
6
Q
line graph
most common ABA graph
A
- based on the cartesian plane
- examine level, trend, variability
- use when data can be scaled along some dimension: time, order of responses in a sequence
7
Q
7 parts of line graph
A
- Y
- X
- condition lines: solid (major change in IV) & dash (minor change in IV)
- condition labels
- data point
- data path: max. 4 data paths in 1 set of axes
- figure capture
8
Q
bar graph
A
- based on the cartesian plane
- NO distinct data points representing successive response measures through time
- can NOT be used with time
- NOT allow for variability & trend analysis
- use when:
- show separate sets of data that are NOT related
- summarize performance within a condition / a group of individuals
9
Q
cumulative records
- develop by skinner in EAB research in 1957
A
- cumulative recorder enables a subject to draw one’s own graph
- when the total # of responses exceed the upper limit of the Y-axis, the data path resets to ZERO on the Y-axis & begins its rise again
- used for rate/frequency data
10
Q
2 types cumulative record response rates
A
- overall response rates: average rate of response over a given time period = total # of response / # of observation periods
- local response rate: average rate of response during periods of time smaller than that for which an overall response rate has been given = total # of increased response during specified periods / # of observation periods specified
11
Q
semilogarithmic charts
A
- logarithmic scales look at B change through PROPORTIONAL / RELATIVE change
- semilogarithmic: 1 axis is scaled proportionally
- x: equal intervals
- y: scaled logarithmic
- data that shown as an exponential curve on an equal interval chart is a straight line on a semilogarithmic chart
12
Q
standard celeration chart
A
- semilogarithmic charts
- Ogden Lindsley created to be used in Precision Teaching
- to provided a standardized means of charting & analyzing how freq changes over time
- scales go up by MULTIPLES, as in 2, 4, 8, 16, 32 or 10., 100, 1000
- students SELF-MINITOR their progress by recording data that makes a graph that displays # of items correct & # of errors within fixed periods of time distributed across the day / week
- goal: to INCREASE # of correct & DECREASE # of errors within the set time –> promoting rate of responding
- semilogarithmic chart allows data to be squeezed into progressively tighter & tighter bundles
- whenever B changes within a given time period, the slope of the change looks the SAME whether you start with a very high or low level of B
13
Q
scatter plots
A
- show the relative distribution of individual measures in a data set
- data points are unconnected
- used to show the temporal distribution / time of the behavior
- the grouping of the individual data points may help to identify elusive environmental stimuli
14
Q
3 fundamental properties of B change
A
- level
- trend
- variability
15
Q
level
A
- value on the vertical axis around which a series of data measures converge
- change in level: when the data’s average value changes
- can be examined by looking at mean, median, range
- mean level line
- median level line
