section 7-B data

Question 1

data

Answer 1

• medium with which the behavior analyst works
• results of measurement
• empirical basis for decision making

Question 2

graphs

Answer 2

• visual format for displaying data
• reveal relations between measurements & variables
• how BA organize, store, interpret, and communicate the results of our work

Question 3

3 purpose of graphs

Answer 3

• communicate our data
• assess data
• show how DV/IV related

Question 4

5 benefits of graphs

Answer 4

• 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

Question 5

5 types graphs

Answer 5

equal-interval graphs:

• line
• bar
• cumulative record
• scatter plot

non-equal interval graphs:
- standard celebration chart (logarithmic scales)

Question 6

line graph

most common ABA graph

Answer 6

• 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

Question 7

7 parts of line graph

Answer 7

1. Y
2. X
3. condition lines: solid (major change in IV) & dash (minor change in IV)
4. condition labels
5. data point
6. data path: max. 4 data paths in 1 set of axes
7. figure capture

Question 8

bar graph

Answer 8

• 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:
  
  1. show separate sets of data that are NOT related
  2. summarize performance within a condition / a group of individuals

Question 9

cumulative records

- develop by skinner in EAB research in 1957

Answer 9

• 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

Question 10

2 types cumulative record response rates

Answer 10

1. overall response rates: average rate of response over a given time period = total # of response / # of observation periods
2. 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

Question 11

semilogarithmic charts

Answer 11

• 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

Question 12

standard celeration chart

Answer 12

• 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

Question 13

scatter plots

Answer 13

• 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

Question 14

3 fundamental properties of B change

Answer 14

• level
• trend
• variability

Question 15

level

Answer 15

• 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

Question 16

trend

Answer 16

• overall direction taken by the data path
• the general direction & rate of increase or decrease over time
• trend line/line of progress: straight line drawn through the data to show the trend
• equal # of data points fall above & below the line
• method to draw trend line
  
  1. freehand
  2. linear regression equation
  3. SPLIT- MIDDLE LINE OF PROGRESS

Question 17

6 steps to a split-middle line of progress

Answer 17

1. count: # of data points
2. divide: draw a vertical line to divide # of data points in half
3. mid-rate: for each half of the data, find the middle value on Y-axis
4. mid-date: for each half of the data, find the middle point on x-axis
5. quarter-intersect line of progress: connect the 2 mid-rate & mid-date points of intersection
6. split-middle line of progress: shift the quarterly-intersect line up or down so that an equal # of points fall above & below it.

Question 18

variability

Answer 18

• frequency & degree to which multiple measures of B yield different outcomes
• high degree of variability = little / no control over the factors influencing B

Question 19

visual analysis of temporal relations of data within & between conditions

Answer 19

• visual analysis of temporal relations of data within conditions
• visual analysis of temporal relations of data between conditions