Module 7 lab: observing behavior and applying single-subject research designs to change behavior

Make sure you read chapter 13 sections on describing and measuring behavior (yest, that was module 6) and all of chapter 14. It will also be helpful for you to review the example (below) before completing the lab.

One description of the purpose of psychology is to “describe, predict, and influence behavior.”

Observational and correlational designs are very sueful for describing behavior andSingle case designs are a perfect for predicting and evaluating the effectiveness of out attempts to influence the behavior of an individual.

Select a behavior of someone (or something, for example your pet) and how you could develop a series of single-subject designs to describe, predict and influence the behavior of interest. You might apply the behavior change strategy to a friend or family member, a roommate or neighbor, employer or employee, a teammate or classmate, even a pet – anyone whose behavior you would like to change.

You can use this worksheet to complete the lab. Module 7 lab.docx Download Module 7 lab.docx

A1. In a few sentences, describe the behavior you want to change.

A2. Operationalize the outcome you want. For instance, if the behavior of interest is ‘cigarette smoking’ you might want to see it decrease to ‘zero cigarettes per day’. On the other hand, maybe you are okay with a reduction in smoking of 50% or more, or perhaps even getting the smoker to smoke outside instead of in your home, and you could operationalize as ‘number of cigarettes smoked indoors per day.’

Make sure your outcome is an observable behavior. For example, if weight reduction was a goal, ‘weight’ is not a behavior. Instead you might operationalize it as ‘calories consumed per day; or perhaps as ‘servings of fruit and vegetables eaten per day,’ or ‘minutes of walking per day.’

B. Describe how you would use behavioral observation methods to be able to describe the behavior in detail. For instance to understand how often or in what situations the behavior occurs and to identify variables that might affect the frequency, duration of the behavior and factors that might be related to increases or decreases in the behavior. Specifically:

B 1. (see chapter 13 on behavioral observations for answering the B questions) How will you measure the behavior of interest? Describe it, e.g., I will ask the person how many cigaretters they smoke each day.  I will monitor my roommate and track how many times a week, (or how many minutes a day) they clean house, operationalized as washing dishes, cleaning the kitchen counter and sinks, cleaning the bathroom, dusting the living room, or sweeping/mopping the floors.

B 2. Is your measurement method a frequency, duration, or interval method?

B3. Why did you choose your measurement method instead of one of the other two measurement methods?

B4. What sort of outcome would you expect to see given your observations. That is, what variables do you think influence the behavior of interest, or appear to cause the behavior to increase of descrease?

For instance, perhaps your roommate does less housework on weeks she has an exam, or a relative appears to drink more when he describes himself as “stressed out,” or your cat ‘sprays’ the furniture when you leave him alone for more than 12 hours, etc.

C1. Describe your intervention or treatment.  It could be punishing ‘bad’ behavior, offering rewards or incentives for desired behavior, participation in a treatment program, or a change in the environment (also called ‘stimulus control’). Be specific, operationalize it carefully, for example: a response cost, “taking away the child’s access to video games for one hour each time he swears”, or a reward “paying the participants $5 each day he doesn’t smoke,” or a social reinforce “smiling at the child each time he makes eye contact.” Or “thanking my husband every time he takes out the garbage.” Or an intervention such as“ watching a 10 minute video on the benefits of eating fruits and vegetables every day for a week” Or self-monitoring, “monitoring nail biting every hour for a week.”

An ABAB design, where we measure a behavior at baseline (A) implement an intervention (B) withdraw the intervention and return to baseline (A) and then add the intervention (B) is the best way to show that the intervention was the cause of the behavior change.

C2. How long would your phases need to be? Why?

C3. Describe the trend you would expect to see in each phase of your single subject design.

C4. Why do you think the trend reflected in your selection for C1 will most resemble your outcome? Be very specific. To answer this question make sure you understand the sections in chapter 14 on ‘trend, level, and stability’ and “the ABAB reversal design” and its limitations.

C5. What is one limitation of your design

D.1 If you were to use a more complex design which of the following would you select?

Multiple baseline design, component analysis, changing criterion, alternating treatments

I would use a changing criterion design

D2. Briefly describe the variables in your more complex design.

D3. Why would this design enhance the knowledge you gained from the ABAB design?

Sample completed lab

A1. In a few sentences, describe the behavior you want to change.

I want to stop my dog from getting dirt all over my floors. My dog Franklin (aka Frankie, Big Boy, and Mr. Good Boy) has recently been using his big nose to toss dirt out of one of my houseplants onto the floor.  He does this at least once a day. If I move the plant he finds another one, and I don’t want to remove all of my plants.

A2. Operationalize the outcome you want. For instance, if the behavior of interest is ‘cigarette smoking’ you might want to see it decrease to ‘zero cigarettes per day’. On the other hand, maybe you are okay with a reduction in smoking of 50% or more, or perhaps even getting the smoker to smoke outside instead of in your home, and you could operationalize as ‘number of cigarettes smoked indoors per day.’

Make sure your outcome is an observable behavior. For example, if weight reduction was a goal, ‘weight’ is not a behavior. Instead you might operationalize it as ‘calories consumed per day; or perhaps as ‘servings of fruit and vegetables eaten per day,’ or ‘minutes of walking per day.’

I want the behavior to stop, that is, a frequency of zero operationalized as my dog will not dig in any houseplant during one week of observation.

B. Describe how you would use behavioral observation methods to be able to describe the behavior in detail. For instance to understand how often or in what situations the behavior occurs and to identify variables that might affect the frequency, duration of the behavior and factors that might be related to increases or decreases in the behavior. Specifically:

B 1. (see chapter 13 on behavioral observations for answering the B questions) How will you measure the behavior of interest? Describe it, e.g., I will ask the person how many cigarettes they smoke each day.  I will monitor my roommate and track how many times a week, (or how many minutes a day) they clean house, operationalized as washing dishes, cleaning the kitchen counter and sinks, cleaning the bathroom, dusting the living room, or sweeping/mopping the floors.

I will use a video camera placed near the plant and observe how many times a day he digs dirt out of the plant. When possible I will observe what he was doing just before he dug in the plant.

B 2. Is your measurement method a frequency, duration, or interval method?

Frequency method

B3. Why did you choose your measurement method instead of one of the other two measurement methods?

The behavior is not suited for measuring ‘duration’ because the duration is always the same and it just a second or so. It does not occur frequently enough for an interval to make sense. The interval would be too long to be practical.

B4. What sort of outcome would you expect to see given your observations? That is, what is the expected frequency, duration, or interval of
the behavior? What variables do you think influence the behavior of interest, or appear to cause the behavior to increase of decrease?

For instance, perhaps your roommate does less housework on weeks she has an exam, or a relative appears to drink more when he describes himself as “stressed out,” or your cat ‘sprays’ the furniture when you leave him alone for more than 12 hours, etc.

He would dig in the plants about once or twice a day. I think he does this behavior more when I am at home than when I am not at home, and that he does it when he wants my attention and I am working.

C1. Describe your intervention or treatment.  It could be punishing ‘bad’ behavior, offering rewards or incentives for desired behavior, participation in a treatment program, or a change in the environment (also called ‘stimulus control’). Be specific, operationalize it carefully, for example: a response cost, “taking away the child’s access to video games for one hour each time he swears”, or a reward “paying the participants $5 each day he doesn’t smoke,” or a social reinforce “smiling at the child each time he makes eye contact.” Or “thanking my husband every time he takes out the garbage.” Or an intervention such as“ watching a 10 minute video on the benefits of eating fruits and vegetables every day for a week” Or self-monitoring, “monitoring nail biting every hour for a week.”

An ABAB design, where we (A) measure a behavior at baseline, and then (B) implement an intervention, then (A) withdraw the intervention and return to baseline, and finally (B)  resume the intervention is the best way to show that the intervention was the cause of the behavior change.

My ‘treatment’ will be to give my dog at least five minutes of attention every hour when I am home. He gets lots of attention every day, and the strategy is to make it more frequent rather than necessarily longer in total weekly duration. (Note: other options might be to punish him every time I catch him near the plant by reprimanding him or squirting him with water. Or giving him new toys in case the behavior is caused by boredom rather than attention)

C2. How long would your phases need to be? Why?

My phases will last a week with one data point (frequency of digging in plant) on each day, and it might be possible to do it in five days of observation. It is a low frequency behavior, so I would need several days for each phase to see that the data are stable. (Note: depending on the frequency and duration of the behavior of interest your phases could be much longer or shorter and there is a minimum number of data points needed to observe a trend)

C3.Describe the trend you would expect to see in each phase of your single subject design.

In the first A phase with daily monitoring I would expect to see the behavior occur once or twice a day.  In the first B phase I would expect the behavior to decline to zero. I think it would revert back to once or twice a day when I return to baseline, and then reduce to zero again when I implement the intervention.

Note: this example is simple case where the behavior completely decreases and then increases right away. For other behaviors you might see more variability in the baseline phase, or see it gradually increase or decrease in the B phase. It might gradually or immediately return to baseline, and if the intervention has lasting effects it might not revert to baseline at all (for instance, one type of treatment might have a lasting effect, while another needs to be administered daily). Depending on your intervention the increase or decrease in behavior in the second B phase might be immediate or gradual.)

C4. Why do you think the trend reflected in your selection for C1 will most resemble your outcome? Be very specific. To answer this question make sure you understand the sections in chapter 14 on ‘trend, level, and stability’ and “the ABAB reversal design” and its limitations.

The behavior is low frequency to begin with and if it is being caused by attention-seeking and occurs only once or twice a day then providing attention every hour should decrease it quickly. Returning to baseline will lead to immediate increase in the behavior since the dog will no longer be getting attention and attention doesn’t have very long lasting effects, but when the attention is resumed the behavior should increase.

C5. What is one limitation of your design?

A few limitations are (1) the behavior is low frequency, so there isn’t much room for improvement (2) (2).  My observation plan doesn’t allow me to be sure that the behavior is maintained by attention. If something other than attention is causing the behavior, then the behavior will not change after the intervention. (3) On some days I am home more than others so the total attention the dog gets will vary from day to day.

D.1 If you were to use a more complex design which of the following would you select and why?

I would use an alternating treatments design since I am not 100% sure that the behavior is influenced Multiple baseline design, component analysis, changing criterion, alternating treatments

D2. Briefly describe the variables in your more complex design.

 I would alternate attention with punishment, that is, on each day I would flip a coin and depending on heads or tails either increase attention frequency to once per hour or squirting him with a water spray bottle each time he digs in the plant. (Note: since I have two dogs another option for D1 and D2 might be to use a multiple baseline design across subjects. With a more complex treatment I might use a component analysis, and with a more frequent behavior I might use a changing criterion design, See chapter 14 for the rationale for each)

D3. Why would this design enhance the knowledge you gained from the ABAB design?

The alternating treatments design would allow me to simultaneously evaluate the effectiveness of two treatments and I could see if one is better than the other.