Chapter 3 Graphing Behavior and Measuring Change
Chap t e r 3 Graphi n g Beh a v i o r and Mea s u ri n g Chan g e Chap t e r Outlin e Compo n e n t s of a Grap h Y-axis and X-axis Axis Labels Axis Num b e r s Data Points Phas e Lines Phas e Labels Grap hi n g Beha vior al Data Grap hi n g Dat a from Differe n t Recordi n g Proce d u r e s Rese a r c h Design A-B Design A-B-A-B Design Multiple Baselin e Design Across Subject s Across Beh aviors Across Settings Altern a ti n g- Treat m e n t s Design Chan gi n g Criterion Design Chapt e r Sum m a r y Key Term s Practic e Test Application s Misap plicatio n s Chapt e r 3, Quiz 1 Chapt e r 3, Quiz 2, Chapt e r 3, Quiz 3
Ide a s for Clas s Activi ti e s 1. Make up a set of dat a with obs erv a ti o n s for a targe t beh a vior in bas elin e and tre at m e n t condition s, provid e the dat a set to the stud e n t s , and hav e the stud e n t s com p o s e a gra p h of the dat a. You can mak e the dat a set simpl e to reflect an A-B design or more com pl e x to reflect an A-B-A-B or multiple bas elin e design. Ask the stud e n t s what desi gn is illustrat e d in their gra p h. 2. Make up hypot h e ti c al res e a r c h sce n a rio s and ask the stud e n t s how they would apply the differe n t res e a r c h design s to the scen a ri os. For ex a m pl e , “Sup p o s e you hav e two presc h o ol stud e n t s who eng a g e in disru ptiv e beh a vi or in the classro o m and you want to evalu a t e an interv e n ti o n to decr e a s e the disru ptiv e beh a vi or. Describ e how you would use a multiple bas elin e acros s subj ect s desi gn to eval u a t e the interv e n ti o n in this sce n a rio.”
Ans w e r s to Prac ti c e Tes t Qu e s t i o n s
23
1. A gra p h is use d to docu m e n t whet h e r a beh a vior has cha n g e d when a beh a vi or modification proc e d u r e is imple m e n t e d . The gra p h mak e s it easi er to com p a r e the levels of the beh a vi or before, during, and after trea t m e n t bec a u s e the levels are visually pres e n t e d for com p a ri s o n . 2. The two varia bl e s illustrat e d in a beh a vi or modificatio n grap h are tim e and beh a vi or . 3. The Y axis, or ordina t e , is the vertical axis of the grap h. The X axis, or absci ss a, is the horizont al axis of the gra p h. The two axe s com e toge t h e r at the bott o m left of the pag e . 4. The level of beh a vior is indicat e d on the Y axis. Time is indicat e d on the X axis. 5. A pha s e is a period of tim e in which the sa m e trea t m e n t (or no trea t m e n t ) is in effect. 6. Dat a point s are not conn e c t e d acros s pha s e lines so that differe n c e s in the level of the beh a vi or in differe n t pha s e s can be se e n mor e easily. 7. Your grap h shoul d includ e: (1) the X axis and the Y axis; (2) the label for the X axis and Y axis; (3) the num b e r s on the Y axis and the X axis; (4) dat a point s; (5) pha s e line(s); and (6) pha s e labels. 8. The label for the Y axis of a grap h bas e d on interv al recordi n g will be the “perc e n t a g e of interv al s” in which the targ e t beh a vi or occurr e d . 9. The A-B design is the sim pl e s t type of design use d in beh a vi or modificatio n, havin g just two pha s e s : bas elin e and trea t m e n t . A = bas elin e; B = trea t m e n t . 10. In the A-B-A-B revers al desi gn, bas elin e and tre at m e n t pha s e s are imple m e n t e d twice. After the first trea t m e n t pha s e , the res e a r c h e r rem o v e s the tre at m e n t and revers e s back to bas elin e. This seco n d bas elin e is then followe d by replication of the trea t m e n t . See Figure 3.11 for an ex a m pl e of an A-B-A-B rever s al desi gn. 11. A multipl e bas elin e desi gn is a rese a r c h design in which a trea t m e n t is evalu a t e d as it is applied seq u e n t i ally to two or mor e subject s, two or mor e beh a vi or s, or in two or mor e setting s. There are thre e typ e s of multipl e bas elin e design s. (1) In a multiple bas elin e acros s subjec t s , ther e is a bas elin e and a tre at m e n t pha s e for the sa m e targe t beh a vior of two or mor e differe n t subj ect s (Figure 3.12). (2) In a multiple bas elin e acros s beh a vi or s , ther e is a bas elin e and trea t m e n t pha s e for two or more differe n t beh a vior s of the sa m e subj ect (Figur e 3.13). (3) In a multipl e bas elin e acros s settin gs , ther e is a bas elin e and trea t m e n t pha s e for two or more settin gs in which the sa m e beh a vior of the sa m e subject is m e a s u r e d (Figure 3.14). 12. An extra n e o u s variabl e is any variabl e that is not being m e a s u r e d or ma ni p ul a t e d as part of trea t m e n t . By withdr awi n g and the n re- imple m e n t i n g tre at m e n t , it is possi bl e to de m o n s t r a t e that beh a vior cha n g e coincid e s with the applicatio n of trea t m e n t . This being the cas e, it is rea s o n a b l e to conclud e that beh a vior chan g e is the result of trea t m e n t rath e r tha n som e extra n e o u s variabl e. 13. When trea t m e n t is imple m e n t e d at differe n t tim e s or after a differe n t num b e r of bas elin e dat a point s in a multipl e bas elin e desi gn, we say that tre at m e n t is “st a g g e r e d ” over tim e. 14. In an alter n a ti n g tre at m e n t s desi gn (ATD) the bas elin e and tre at m e n t condition s (or
24
two tre at m e n t condition s) are cond u c t e d in rapid succ e s si o n and com p a r e d to eac h other (see Figure 3.15). 15. To judge the effective n e s s of trea t m e n t in an ATD, the level of the beh a vi or in the two condition s (bas elin e and tre at m e n t or two differe n t tre at m e n t s ) are com p a r e d within the sa m e tim e period. 16. A cha n gi n g criterion desi gn typically includ e s a bas elin e and trea t m e n t pha s e , with seq u e n t i al perfor m a n c e criteri a specified within the tre at m e n t pha s e (se e Figure 3.16).
25
17. The effectiv e n e s s of trea t m e n t in a cha n gi n g criterion design is det er m i n e d by whet h e r the subj ect’s beh a vior or perfor m a n c e chan g e s to m e e t the chan gi n g perfor m a n c e criteri a. 18. A function al relation s hi p betw e e n a beh a vior modification proc e d u r e and a targe t beh a vi or exist s whe n the targ e t beh a vi or chan g e s as a function of the proc e d u r e (the proc e d u r e caus e s the beh a vi or to chan g e ). You det er m i n e that a function al relation s hi p exist s when the targe t beh a vior cha n g e s if and only if the proce d u r e is imple m e n t e d and the proc e s s is rep e a t e d one or mor e tim e s.
Answers to Applications 2. See Figure 3.18. 3. See Figure 3.19.
Answers to Misapplications 1. The problem with the A-B-A-B design in this example is that, if Ace Consultants reverse the intervention and remove the incentive system, the workers may slow down their work rate and the company may lose money again. This would be damaging to the company and thus the use of the A-B-A-B design in this case would be inappropriate. It would be better to simply keep the incentive system in place and measure the workers' behavior over a long period of time so as to determine whether the productivity stays at the high level. The company is more interested in the change in the workers' productivity than in proving that the incentive system caused the change. 2. The problem in this example is that Alice is plotting the running distance on her graph only once a week. In a self-management program, the individual should monitor the target behavior and plot the behavior on the graph on a daily basis. The individual is less likely to forget if plotting the behavior is a daily routine. In addition, once the intervention is implemented, daily plotting of the target behavior will function as a daily reminder to engage in the behavior. 3. The mistake that Dr. Pete made was to implement treatment for all three target behaviors at the same time. In a multiple-baseline-across-behaviors design, the intervention with each of the individual's target behaviors is staggered over time. In this case, Dr. Pete should implement the intervention with one behavior (for example, initiating conversations) while the other two are still in baseline. Then, after the first behavior improves, he should implement treatment with the second behavior (for example, answering questions) while the third behavior is still in baseline. When the second behavior changes as a result of the intervention, he should implement the intervention with the third behavior (smiling), to produce a change in that behavior also.
Ans w e r s to Quizz e s Quiz 1 1. gra p h 2. Y 3. X 4. bas elin e, trea t m e n t 5. tre at m e n t is not replicat e d 6. A-B-A-B desi gn 7. settin gs 8. chan gi n g criterion desi gn 9. subj ect s 10. beh a vi or s Quiz 2 1. the level of beh a vior 2. the units of tim e bas elin e, trea t m e n t 4. function al relation s hi p 5. A-B-A-B rever s al 6. sta g g e r e d 7. alter n a ti n g tre at m e n t s 8. beh a vi ors 9. setting s 10. A-B Quiz 3 1. beh a vior 2. days, can s of sod a 3. A-B desi gn 4. function al relation s hi p 5. is not 6. A-B-A-B rever s al 7. A-B-A-B rever s al 8. multipl e bas elin e acros s beh a vi ors 9. multipl e bas elin e acros s settin gs 10. multiple bas elin e acros s subj ect s
26
Tes t Qu e s t i o n s 1. In behavior modification, the main way to document behavior change is through the use of a(n): a) graph b) questionnaire c) interview d) statistical analysis ANS: A 2. The two variables represented on a graph are: a) the independent and dependent variables b) time and behavior c) the extraneous and controlling variables d) frequency and outcome ANS: B 3. A data point on a graph tells you: a) when the behavior was recorded b) the level of behavior c) what is causing the behavior d) A and B ANS: D 4. The horizontal axis in a graph is used to record: a) behavior b) time c) change d) duration ANS: B 5. The level of the target behavior is recorded on the: a) horizontal axis b) vertical axis c) x-axis d) abscissa ANS: B 6. There are ________ components necessary for a graph to be complete. a) four b) six c) eight d) ten ANS: B 7. On a graph ___________ is indicated on the Y axis and ________________ is indicated on 27
the X axis. a) time, behavior b) behavior, time c) frequency, duration d) time, treatment ANS: B 8. Which of the following is NOT necessary for a graph to be complete? a) phase lines b) data points c) x-axis and y-axis d) behavior change ANS: D 9. A vertical line on a graph indicating a change in treatment is called a: a) baseline b) time line c) phase line d) treatment line ANS: C 10. Which dimension(s) of behavior can be graphed? a) frequency b) duration c) intensity d) all of these ANS: D 11. In behavior modification, a research design is used to: a) determine if the treatment changed the behavior b) control the target behavior c) make it easier to change the behavior d) A and B ANS: A 12. Billy has trouble sitting still in class, so his teacher implements a program in which she praises Billy for staying in his seat. Billy’s sitting behavior increases, and the teacher concludes her praise was responsible. The teacher does not realize Billy has started taking medication that causes fatigue. The medication would be an example of a(n) __________ variable. a) independent b) extraneous c) dependent d) behavioral ANS: B 13. An A-B design does not demonstrate a functional relationship between the treatment and the 28
target behavior because there is no: a) change in the target behavior b) statistical test of significance c) replication d) validation ANS: C
29
14. A researcher demonstrates that receiving gold stickers for completing math problems increases the number of problems that a student finishes. The researcher has demonstrated a _____________ relationship between receiving gold stickers and completing math problems. a) correlational b) functional c) perfect d) two-way ANS: B 15. In behavior modification, a research design involves: a) estimation b) treatment implementation c) replication d) B and C ANS: D 16. Which of the following is NOT characteristic of an A-B design? a) it is used to compare baseline and treatment b) it is used in applied settings c) it demonstrates a functional relationship d) it is the simplest type of design ANS: C 17. A teacher records the biting behavior of a child before and after treatment is implemented. This is an example of a(n) ____________ design. a) A-B-A-B b) A-B c) Multiple-baseline d) Alternating treatments ANS: B 18. Which research design is referred to as a reversal design? a) A-B b) A-B-A-B c) Multiple-baseline d) Changing-criterion ANS: B 19. In an A-B-A-B design, B refers to ______________ and A refers to _________________. a) baseline; assessment b) baseline; treatment c) treatment; baseline d) dependent variable; independent variable ANS: C 20. What should be taken into account before using an A-B-A-B design? 30
a) Is it ethical to remove treatment? b) Can treatment be removed? c) Will the level of behavior reverse? d) all of these ANS: D 21. Which of the following is a type of multiple-baseline design? a) across-subjects b) across- staff c) across-time d) all of these ANS: A 22. A researcher records the hair pulling behavior of three different children in a classroom and implements treatment at a different time for each child. The researcher is using a(n) _____________ design. a) multiple-baseline-across-behaviors b) multiple-baseline-across-subjects c) alternating-treatments d) changing-criterion ANS: B 23. A researcher is interested in measuring the hitting and screaming behaviors of a child, treating the hitting first, and treating the screaming later. To do this, the researcher would use a(n) ____________ design. a) multiple-baseline-across-settings b) multiple-baseline-across-behaviors c) A-B-A-B d) alternating treatments ANS: B 24. A researcher records a child’s crying behavior when the child is at school and when the child is at home and provides treatment at school then at home. The researcher is using a(n) _____________ design. a) A-B-A-B b) multiple-baseline-across-settings c) changing-criterion d) multiple-baseline-across-behaviors ANS: B 25. A critical feature of multiple-baseline designs is that: a) baselines differ in length b) treatment is staggered over time c) A and B d) none of these ANS: C 31
26. In a(n) _____________ research design, baseline and treatment conditions are conducted in rapid succession and compared to each other. a) A-B-A-B b) multiple-baseline c) changing-criterion d) alternating-treatments
32
27. The research design in which there are increasing (or decreasing) goal levels for the target behavior is referred to as a(n) ____________ design. a) alternating-treatments b) changing-criterion c) multiple-baseline d) A-B-A-B ANS: B 28. Which research design does NOT control for the effect of confounding variables? a) A-B b) A-B-A-B c) changing-criterion d) multiple-baseline ANS: A 29. To det er mi n e whet h e r a new drug decr e a s e s self- injurious beh a vior, the res e a r c h e r me a s u r e s self- injurious beh a vior while giving a subject the drug on eve n num b e r e d days and giving a plac e b o on odd nu mb e r e d days for two weeks. This is an exa m pl e of a(n) ____________________ design. a) multiple bas elin e acros s setting s b) chan gin g criterion c) multiple bas elin e acros s beh a viors d) alter n a tin g tre a t m e n t s ANS: D 30. To det er mi n e whet h e r a new drug decr e a s e s self- injurious beh a vior, the res e a r c h e r me a s u r e s the self- injurious beh a vior of 3 children, and the n gives the drug to the first child after one week. After two weeks of bas elin e, the seco n d child gets the drug and after 3 weeks of bas elin e, the third child gets the drug. This is an exa m pl e of a(n)___________ design. a) multiple bas elin e acros s subject s b) chan gin g criterion c) multiple bas elin e acros s beh a viors d) alter n a tin g tre a t m e n t s ANS: A 1. A ________ is a visual repr e s e n t a ti o n of the occurr e n c e of a beh a vior over time. ANS: graph 2. On a grap h in beh a vior modification, the _____ axis shows the level of the beh a vior. ANS: Y 3. On a grap h in beh a vior modification, the _____ axis shows the units of time. 33
ANS: X 4. In an A-B design, A= _________ and B= __________. ANS: baseline; treatment 5. In a multiple bas elin e acros s subject s design, ther e is a bas elin e and tre a t m e n t pha s e for __________ beh a vior(s) of __________ subject(s). ANS: one; two or more
34
6. In a multiple bas elin e acros s beh a viors design, ther e is a bas elin e and tre a t m e n t pha s e for __________ beh a vior(s) of __________ subject(s). ANS: two or more; one 7. In a multiple bas elin e acros s settin g s design, ther e is a bas elin e and tre a t m e n t pha s e for ________ beh a vior(s) of _________ subject(s) in __________ settin g(s). ANS: one; one; two or mor e 8. In a __________________ res e a r c h design, bas elin e and tre a t m e n t ses sion s occur in rapid succe s sio n. ANS: alternating treatment 9. In a(n) __________________ res e a r c h design, ther e is a bas elin e pha s e followe d by a tre a t m e n t pha s e and the n anot h e r bas elin e and tre a t m e n t phas e . ANS: A-B-A-B 10. In a(n) ___________________ res e a r c h design, ther e is a bas elin e and a tre a t m e n t pha s e and, in the tre a t m e n t pha s e , ther e are differe n t criterion levels for the beh a vior. ANS: changing criterion 1. T F The units of beh a vior are labele d on the Y axis of a grap h. ANS: T 2.
T F Data points should be conn e c t e d acros s pha s e lines.
ANS: F 3.
T F A-B design s are frequ e n tly used in beh a vior modification res e a r c h .
ANS: F 4. T F In the A-B-A-B revers al design, bas elin e and tre a t m e n t are imple m e n t e d twice for one beh a vior of one subject. ANS: T 5.
T F Phas e labels on a grap h are typically put below the X axis.
ANS: F 6. T F In a multiple bas elin e acros s beh a viors design, tre a t m e n t is first imple m e n t e d for one beh a vior of the subject and is then imple m e n t e d later for a seco n d beh a vior of the sa m e subject. ANS: T 7.
T F A function al relations hip betw e e n the tre a t m e n t and the beh a vior 35
is de mo n s t r a t e d in a multiple bas elin e acros s subject s design when the beh a vior of each subject chan g e s only after the tre a t m e n t is imple m e n t e d for that subject. ANS: T 8. T F In an altern a tin g tre a t m e n t s design, the bas elin e is imple m e n t e d for a long period of time and the n the tre a t m e n t is imple m e n t e d for a long period of time. ANS: F 9. T F Each dat a point on a grap h shows you the level of the beh a vior at a particular point in time. ANS: T 10. T F When you de m o n s t r a t e that a beh a vior modification proc e d u r e caus e d a beh a vior to chan g e , you are de m o n s t r a ti n g a function al relation s hip. ANS: T 1. Prior to tre a t m e n t , Jenny smok e d the following num b e r of cigar e t t e s on cons e c u tiv e days; 15, 15, 16, 14, 20, 15, 16, 17, 14, 20. After tre a t m e n t , Jenny smok e s the following num b e r of cigar e t t e s on cons e c u tiv e days; 10, 10, 8, 8, 6, 5, 4, 2, 2, 0, 0, 0, 0. Draw the grap h that show s the frequ e n c y of Jenny’s cigar e t t e smoking before and after tre a t m e n t . Be sure to includ e all of the ess e n ti al comp o n e n t s of a compl e t e grap h. ANS: See Figure 3.10 for an example of an A-B design. 2. Does the A-B design de m o n s t r a t e a function al relation s hip? Why or why not? ANS: No . You determine that a functional relationship exists when the target behavior changes if and only if the procedure is implemented and the process is repeated one or more times. There is no replication in an AB design. 3. Draw an exa m pl e of a grap h of an A-B-A-B rever s al design. ANS: See Figure 3.11 for an example of an A-B-A-B reversal design. 4. What does it me a n to say that tre a t m e n t is sta g g e r e d in a multiple bas elin e design? ANS: When treatment is implemented at different times in a multiple-baseline design, we say that treatment is “staggered” over time. 5. Dr. Allen evalu a t e d the effects of the drug, Ritalin, on the aggr e s siv e beh a vior of a 6- year- old boy using an altern a tin g tre a t m e n t s design. He me a s u r e d the num b e r of aggr e s siv e beh a viors (hits) each day. He alter n a t e d days in which the drug was given with days in which the drug 36
was not given. The drug was given on eve n num b e r e d days and the drug was not given on odd num b e r e d days. The results were as follows. Days
1
2
3
4
5
6
7
8
9
10
1 1
1 2
# of Hits
12
5
15
6
13
2
11
4
15
3
1 2
4
Draw the alter n a ti n g tre a t m e n t s design grap h of thes e results. ANS: See Figure 3.15 for an alternating treatments design.
37
Ide a s for Clas s Activi ti e s 1. Make up a set of dat a with obs erv a ti o n s for a targe t beh a vior in bas elin e and tre at m e n t condition s, provid e the dat a set to the stud e n t s , and hav e the stud e n t s com p o s e a gra p h of the dat a. You can mak e the dat a set simpl e to reflect an A-B design or more com pl e x to reflect an A-B-A-B or multiple bas elin e design. Ask the stud e n t s what desi gn is illustrat e d in their gra p h. 2. Make up hypot h e ti c al res e a r c h sce n a rio s and ask the stud e n t s how they would apply the differe n t res e a r c h design s to the scen a ri os. For ex a m pl e , “Sup p o s e you hav e two presc h o ol stud e n t s who eng a g e in disru ptiv e beh a vi or in the classro o m and you want to evalu a t e an interv e n ti o n to decr e a s e the disru ptiv e beh a vi or. Describ e how you would use a multiple bas elin e acros s subj ect s desi gn to eval u a t e the interv e n ti o n in this sce n a rio.”
Ans w e r s to Prac ti c e Tes t Qu e s t i o n s
23
1. A gra p h is use d to docu m e n t whet h e r a beh a vior has cha n g e d when a beh a vi or modification proc e d u r e is imple m e n t e d . The gra p h mak e s it easi er to com p a r e the levels of the beh a vi or before, during, and after trea t m e n t bec a u s e the levels are visually pres e n t e d for com p a ri s o n . 2. The two varia bl e s illustrat e d in a beh a vi or modificatio n grap h are tim e and beh a vi or . 3. The Y axis, or ordina t e , is the vertical axis of the grap h. The X axis, or absci ss a, is the horizont al axis of the gra p h. The two axe s com e toge t h e r at the bott o m left of the pag e . 4. The level of beh a vior is indicat e d on the Y axis. Time is indicat e d on the X axis. 5. A pha s e is a period of tim e in which the sa m e trea t m e n t (or no trea t m e n t ) is in effect. 6. Dat a point s are not conn e c t e d acros s pha s e lines so that differe n c e s in the level of the beh a vi or in differe n t pha s e s can be se e n mor e easily. 7. Your grap h shoul d includ e: (1) the X axis and the Y axis; (2) the label for the X axis and Y axis; (3) the num b e r s on the Y axis and the X axis; (4) dat a point s; (5) pha s e line(s); and (6) pha s e labels. 8. The label for the Y axis of a grap h bas e d on interv al recordi n g will be the “perc e n t a g e of interv al s” in which the targ e t beh a vi or occurr e d . 9. The A-B design is the sim pl e s t type of design use d in beh a vi or modificatio n, havin g just two pha s e s : bas elin e and trea t m e n t . A = bas elin e; B = trea t m e n t . 10. In the A-B-A-B revers al desi gn, bas elin e and tre at m e n t pha s e s are imple m e n t e d twice. After the first trea t m e n t pha s e , the res e a r c h e r rem o v e s the tre at m e n t and revers e s back to bas elin e. This seco n d bas elin e is then followe d by replication of the trea t m e n t . See Figure 3.11 for an ex a m pl e of an A-B-A-B rever s al desi gn. 11. A multipl e bas elin e desi gn is a rese a r c h design in which a trea t m e n t is evalu a t e d as it is applied seq u e n t i ally to two or mor e subject s, two or mor e beh a vi or s, or in two or mor e setting s. There are thre e typ e s of multipl e bas elin e design s. (1) In a multiple bas elin e acros s subjec t s , ther e is a bas elin e and a tre at m e n t pha s e for the sa m e targe t beh a vior of two or mor e differe n t subj ect s (Figure 3.12). (2) In a multiple bas elin e acros s beh a vi or s , ther e is a bas elin e and trea t m e n t pha s e for two or more differe n t beh a vior s of the sa m e subj ect (Figur e 3.13). (3) In a multipl e bas elin e acros s settin gs , ther e is a bas elin e and trea t m e n t pha s e for two or more settin gs in which the sa m e beh a vior of the sa m e subject is m e a s u r e d (Figure 3.14). 12. An extra n e o u s variabl e is any variabl e that is not being m e a s u r e d or ma ni p ul a t e d as part of trea t m e n t . By withdr awi n g and the n re- imple m e n t i n g tre at m e n t , it is possi bl e to de m o n s t r a t e that beh a vior cha n g e coincid e s with the applicatio n of trea t m e n t . This being the cas e, it is rea s o n a b l e to conclud e that beh a vior chan g e is the result of trea t m e n t rath e r tha n som e extra n e o u s variabl e. 13. When trea t m e n t is imple m e n t e d at differe n t tim e s or after a differe n t num b e r of bas elin e dat a point s in a multipl e bas elin e desi gn, we say that tre at m e n t is “st a g g e r e d ” over tim e. 14. In an alter n a ti n g tre at m e n t s desi gn (ATD) the bas elin e and tre at m e n t condition s (or
24
two tre at m e n t condition s) are cond u c t e d in rapid succ e s si o n and com p a r e d to eac h other (see Figure 3.15). 15. To judge the effective n e s s of trea t m e n t in an ATD, the level of the beh a vi or in the two condition s (bas elin e and tre at m e n t or two differe n t tre at m e n t s ) are com p a r e d within the sa m e tim e period. 16. A cha n gi n g criterion desi gn typically includ e s a bas elin e and trea t m e n t pha s e , with seq u e n t i al perfor m a n c e criteri a specified within the tre at m e n t pha s e (se e Figure 3.16).
25
17. The effectiv e n e s s of trea t m e n t in a cha n gi n g criterion design is det er m i n e d by whet h e r the subj ect’s beh a vior or perfor m a n c e chan g e s to m e e t the chan gi n g perfor m a n c e criteri a. 18. A function al relation s hi p betw e e n a beh a vior modification proc e d u r e and a targe t beh a vi or exist s whe n the targ e t beh a vi or chan g e s as a function of the proc e d u r e (the proc e d u r e caus e s the beh a vi or to chan g e ). You det er m i n e that a function al relation s hi p exist s when the targe t beh a vior cha n g e s if and only if the proce d u r e is imple m e n t e d and the proc e s s is rep e a t e d one or mor e tim e s.
Answers to Applications 2. See Figure 3.18. 3. See Figure 3.19.
Answers to Misapplications 1. The problem with the A-B-A-B design in this example is that, if Ace Consultants reverse the intervention and remove the incentive system, the workers may slow down their work rate and the company may lose money again. This would be damaging to the company and thus the use of the A-B-A-B design in this case would be inappropriate. It would be better to simply keep the incentive system in place and measure the workers' behavior over a long period of time so as to determine whether the productivity stays at the high level. The company is more interested in the change in the workers' productivity than in proving that the incentive system caused the change. 2. The problem in this example is that Alice is plotting the running distance on her graph only once a week. In a self-management program, the individual should monitor the target behavior and plot the behavior on the graph on a daily basis. The individual is less likely to forget if plotting the behavior is a daily routine. In addition, once the intervention is implemented, daily plotting of the target behavior will function as a daily reminder to engage in the behavior. 3. The mistake that Dr. Pete made was to implement treatment for all three target behaviors at the same time. In a multiple-baseline-across-behaviors design, the intervention with each of the individual's target behaviors is staggered over time. In this case, Dr. Pete should implement the intervention with one behavior (for example, initiating conversations) while the other two are still in baseline. Then, after the first behavior improves, he should implement treatment with the second behavior (for example, answering questions) while the third behavior is still in baseline. When the second behavior changes as a result of the intervention, he should implement the intervention with the third behavior (smiling), to produce a change in that behavior also.
Ans w e r s to Quizz e s Quiz 1 1. gra p h 2. Y 3. X 4. bas elin e, trea t m e n t 5. tre at m e n t is not replicat e d 6. A-B-A-B desi gn 7. settin gs 8. chan gi n g criterion desi gn 9. subj ect s 10. beh a vi or s Quiz 2 1. the level of beh a vior 2. the units of tim e bas elin e, trea t m e n t 4. function al relation s hi p 5. A-B-A-B rever s al 6. sta g g e r e d 7. alter n a ti n g tre at m e n t s 8. beh a vi ors 9. setting s 10. A-B Quiz 3 1. beh a vior 2. days, can s of sod a 3. A-B desi gn 4. function al relation s hi p 5. is not 6. A-B-A-B rever s al 7. A-B-A-B rever s al 8. multipl e bas elin e acros s beh a vi ors 9. multipl e bas elin e acros s settin gs 10. multiple bas elin e acros s subj ect s
26
Tes t Qu e s t i o n s 1. In behavior modification, the main way to document behavior change is through the use of a(n): a) graph b) questionnaire c) interview d) statistical analysis ANS: A 2. The two variables represented on a graph are: a) the independent and dependent variables b) time and behavior c) the extraneous and controlling variables d) frequency and outcome ANS: B 3. A data point on a graph tells you: a) when the behavior was recorded b) the level of behavior c) what is causing the behavior d) A and B ANS: D 4. The horizontal axis in a graph is used to record: a) behavior b) time c) change d) duration ANS: B 5. The level of the target behavior is recorded on the: a) horizontal axis b) vertical axis c) x-axis d) abscissa ANS: B 6. There are ________ components necessary for a graph to be complete. a) four b) six c) eight d) ten ANS: B 7. On a graph ___________ is indicated on the Y axis and ________________ is indicated on 27
the X axis. a) time, behavior b) behavior, time c) frequency, duration d) time, treatment ANS: B 8. Which of the following is NOT necessary for a graph to be complete? a) phase lines b) data points c) x-axis and y-axis d) behavior change ANS: D 9. A vertical line on a graph indicating a change in treatment is called a: a) baseline b) time line c) phase line d) treatment line ANS: C 10. Which dimension(s) of behavior can be graphed? a) frequency b) duration c) intensity d) all of these ANS: D 11. In behavior modification, a research design is used to: a) determine if the treatment changed the behavior b) control the target behavior c) make it easier to change the behavior d) A and B ANS: A 12. Billy has trouble sitting still in class, so his teacher implements a program in which she praises Billy for staying in his seat. Billy’s sitting behavior increases, and the teacher concludes her praise was responsible. The teacher does not realize Billy has started taking medication that causes fatigue. The medication would be an example of a(n) __________ variable. a) independent b) extraneous c) dependent d) behavioral ANS: B 13. An A-B design does not demonstrate a functional relationship between the treatment and the 28
target behavior because there is no: a) change in the target behavior b) statistical test of significance c) replication d) validation ANS: C
29
14. A researcher demonstrates that receiving gold stickers for completing math problems increases the number of problems that a student finishes. The researcher has demonstrated a _____________ relationship between receiving gold stickers and completing math problems. a) correlational b) functional c) perfect d) two-way ANS: B 15. In behavior modification, a research design involves: a) estimation b) treatment implementation c) replication d) B and C ANS: D 16. Which of the following is NOT characteristic of an A-B design? a) it is used to compare baseline and treatment b) it is used in applied settings c) it demonstrates a functional relationship d) it is the simplest type of design ANS: C 17. A teacher records the biting behavior of a child before and after treatment is implemented. This is an example of a(n) ____________ design. a) A-B-A-B b) A-B c) Multiple-baseline d) Alternating treatments ANS: B 18. Which research design is referred to as a reversal design? a) A-B b) A-B-A-B c) Multiple-baseline d) Changing-criterion ANS: B 19. In an A-B-A-B design, B refers to ______________ and A refers to _________________. a) baseline; assessment b) baseline; treatment c) treatment; baseline d) dependent variable; independent variable ANS: C 20. What should be taken into account before using an A-B-A-B design? 30
a) Is it ethical to remove treatment? b) Can treatment be removed? c) Will the level of behavior reverse? d) all of these ANS: D 21. Which of the following is a type of multiple-baseline design? a) across-subjects b) across- staff c) across-time d) all of these ANS: A 22. A researcher records the hair pulling behavior of three different children in a classroom and implements treatment at a different time for each child. The researcher is using a(n) _____________ design. a) multiple-baseline-across-behaviors b) multiple-baseline-across-subjects c) alternating-treatments d) changing-criterion ANS: B 23. A researcher is interested in measuring the hitting and screaming behaviors of a child, treating the hitting first, and treating the screaming later. To do this, the researcher would use a(n) ____________ design. a) multiple-baseline-across-settings b) multiple-baseline-across-behaviors c) A-B-A-B d) alternating treatments ANS: B 24. A researcher records a child’s crying behavior when the child is at school and when the child is at home and provides treatment at school then at home. The researcher is using a(n) _____________ design. a) A-B-A-B b) multiple-baseline-across-settings c) changing-criterion d) multiple-baseline-across-behaviors ANS: B 25. A critical feature of multiple-baseline designs is that: a) baselines differ in length b) treatment is staggered over time c) A and B d) none of these ANS: C 31
26. In a(n) _____________ research design, baseline and treatment conditions are conducted in rapid succession and compared to each other. a) A-B-A-B b) multiple-baseline c) changing-criterion d) alternating-treatments
32
27. The research design in which there are increasing (or decreasing) goal levels for the target behavior is referred to as a(n) ____________ design. a) alternating-treatments b) changing-criterion c) multiple-baseline d) A-B-A-B ANS: B 28. Which research design does NOT control for the effect of confounding variables? a) A-B b) A-B-A-B c) changing-criterion d) multiple-baseline ANS: A 29. To det er mi n e whet h e r a new drug decr e a s e s self- injurious beh a vior, the res e a r c h e r me a s u r e s self- injurious beh a vior while giving a subject the drug on eve n num b e r e d days and giving a plac e b o on odd nu mb e r e d days for two weeks. This is an exa m pl e of a(n) ____________________ design. a) multiple bas elin e acros s setting s b) chan gin g criterion c) multiple bas elin e acros s beh a viors d) alter n a tin g tre a t m e n t s ANS: D 30. To det er mi n e whet h e r a new drug decr e a s e s self- injurious beh a vior, the res e a r c h e r me a s u r e s the self- injurious beh a vior of 3 children, and the n gives the drug to the first child after one week. After two weeks of bas elin e, the seco n d child gets the drug and after 3 weeks of bas elin e, the third child gets the drug. This is an exa m pl e of a(n)___________ design. a) multiple bas elin e acros s subject s b) chan gin g criterion c) multiple bas elin e acros s beh a viors d) alter n a tin g tre a t m e n t s ANS: A 1. A ________ is a visual repr e s e n t a ti o n of the occurr e n c e of a beh a vior over time. ANS: graph 2. On a grap h in beh a vior modification, the _____ axis shows the level of the beh a vior. ANS: Y 3. On a grap h in beh a vior modification, the _____ axis shows the units of time. 33
ANS: X 4. In an A-B design, A= _________ and B= __________. ANS: baseline; treatment 5. In a multiple bas elin e acros s subject s design, ther e is a bas elin e and tre a t m e n t pha s e for __________ beh a vior(s) of __________ subject(s). ANS: one; two or more
34
6. In a multiple bas elin e acros s beh a viors design, ther e is a bas elin e and tre a t m e n t pha s e for __________ beh a vior(s) of __________ subject(s). ANS: two or more; one 7. In a multiple bas elin e acros s settin g s design, ther e is a bas elin e and tre a t m e n t pha s e for ________ beh a vior(s) of _________ subject(s) in __________ settin g(s). ANS: one; one; two or mor e 8. In a __________________ res e a r c h design, bas elin e and tre a t m e n t ses sion s occur in rapid succe s sio n. ANS: alternating treatment 9. In a(n) __________________ res e a r c h design, ther e is a bas elin e pha s e followe d by a tre a t m e n t pha s e and the n anot h e r bas elin e and tre a t m e n t phas e . ANS: A-B-A-B 10. In a(n) ___________________ res e a r c h design, ther e is a bas elin e and a tre a t m e n t pha s e and, in the tre a t m e n t pha s e , ther e are differe n t criterion levels for the beh a vior. ANS: changing criterion 1. T F The units of beh a vior are labele d on the Y axis of a grap h. ANS: T 2.
T F Data points should be conn e c t e d acros s pha s e lines.
ANS: F 3.
T F A-B design s are frequ e n tly used in beh a vior modification res e a r c h .
ANS: F 4. T F In the A-B-A-B revers al design, bas elin e and tre a t m e n t are imple m e n t e d twice for one beh a vior of one subject. ANS: T 5.
T F Phas e labels on a grap h are typically put below the X axis.
ANS: F 6. T F In a multiple bas elin e acros s beh a viors design, tre a t m e n t is first imple m e n t e d for one beh a vior of the subject and is then imple m e n t e d later for a seco n d beh a vior of the sa m e subject. ANS: T 7.
T F A function al relations hip betw e e n the tre a t m e n t and the beh a vior 35
is de mo n s t r a t e d in a multiple bas elin e acros s subject s design when the beh a vior of each subject chan g e s only after the tre a t m e n t is imple m e n t e d for that subject. ANS: T 8. T F In an altern a tin g tre a t m e n t s design, the bas elin e is imple m e n t e d for a long period of time and the n the tre a t m e n t is imple m e n t e d for a long period of time. ANS: F 9. T F Each dat a point on a grap h shows you the level of the beh a vior at a particular point in time. ANS: T 10. T F When you de m o n s t r a t e that a beh a vior modification proc e d u r e caus e d a beh a vior to chan g e , you are de m o n s t r a ti n g a function al relation s hip. ANS: T 1. Prior to tre a t m e n t , Jenny smok e d the following num b e r of cigar e t t e s on cons e c u tiv e days; 15, 15, 16, 14, 20, 15, 16, 17, 14, 20. After tre a t m e n t , Jenny smok e s the following num b e r of cigar e t t e s on cons e c u tiv e days; 10, 10, 8, 8, 6, 5, 4, 2, 2, 0, 0, 0, 0. Draw the grap h that show s the frequ e n c y of Jenny’s cigar e t t e smoking before and after tre a t m e n t . Be sure to includ e all of the ess e n ti al comp o n e n t s of a compl e t e grap h. ANS: See Figure 3.10 for an example of an A-B design. 2. Does the A-B design de m o n s t r a t e a function al relation s hip? Why or why not? ANS: No . You determine that a functional relationship exists when the target behavior changes if and only if the procedure is implemented and the process is repeated one or more times. There is no replication in an AB design. 3. Draw an exa m pl e of a grap h of an A-B-A-B rever s al design. ANS: See Figure 3.11 for an example of an A-B-A-B reversal design. 4. What does it me a n to say that tre a t m e n t is sta g g e r e d in a multiple bas elin e design? ANS: When treatment is implemented at different times in a multiple-baseline design, we say that treatment is “staggered” over time. 5. Dr. Allen evalu a t e d the effects of the drug, Ritalin, on the aggr e s siv e beh a vior of a 6- year- old boy using an altern a tin g tre a t m e n t s design. He me a s u r e d the num b e r of aggr e s siv e beh a viors (hits) each day. He alter n a t e d days in which the drug was given with days in which the drug 36
was not given. The drug was given on eve n num b e r e d days and the drug was not given on odd num b e r e d days. The results were as follows. Days
1
2
3
4
5
6
7
8
9
10
1 1
1 2
# of Hits
12
5
15
6
13
2
11
4
15
3
1 2
4
Draw the alter n a ti n g tre a t m e n t s design grap h of thes e results. ANS: See Figure 3.15 for an alternating treatments design.
37
