2 - Research methods Flashcards
1
Q
How do you begin a study?
A
- First step: What is your question?
→ Is it clearly defined? What is your hypothesis? - What behavior do you want to measure?
→ Behavior is any activity that can be observed and somehow measured
→ How are you going to measure it?
2
Q
How do we define a behaviour in research?
A
Independent and dependent variables, stimulus and responses, overt and covert behavior, appetitive and aversive stimuli, motivating operations
3
Q
What are variables?
A
- Characteristics of a person, place, or thing that can change (vary) over time or from one situation to another
- Anything that can be defined and measured
- There are Independent variables (IV) and Dependent variables (DV)`
4
Q
What is an independent variable?
A
- Anything that systematically varies across different conditions in an experiment
- It is manipulated or controlled by the experimenter
- These variables can also have levels
→ p.ex: when testing 3 types of drugs, each type will be a level - Research question example: Does the kind of food reward affect the rat’s speed in a maze?
–> IV: The type of food rats in each group receive when they reach their goal
5
Q
What is a dependant variable?
A
- The response/behaviour that is affected by changes in the independent variable
→ It is dependent on changes in the independent variable
→ It is measured by the experimenter - Research question example: Does the kind of food reward affect the rat’s speed in a maze?
–> the speed of the rat to eat all the food reward in the radial arm maze
6
Q
How do IVs and DVs tie into cause-effect functional relationships?
A
- The relationship between changes in the IV (stimulus) and changes in the DV (response, behavior) is the functional relationship
→ Cause(IV)-and-effect(DV) relationship
7
Q
What is a stimulus?
A
- Anything that can potentially influence behaviour
- Also known as a ‘cue’ that signals a behaviour
→ Examples: Food is a stimulus that elicits the response of salivation when presented to a hungry dog
→ Example: A high mark on a test is a stimulus which elicits a response of a grin
8
Q
What is a response?
A
- A particular instance of a behaviour
- The response of one organism can act as a stimulus that stimulates the response of another organism
→ p.ex: when a rat bites another rat, it is a stimulus that elicits the retaliatory response of the second rat to bite back, this is then a stimulus that causes the first rat to retreat which is a response
→ Examples: at a red light (stimulus), we press the brake (response); press the gas
9
Q
How can we differentiate appetitive stimulus from aversive stimulus? Give an example for each
A
-
Appetitive stimulus: An event that an organism will seek out (tend to be pleasant)
→ not necessarily always physiological (but for this class, it will be)
→ Examples: Food when hungry; Water when thirsty -
Aversive stimulus: an event that an organism will avoid (tend to be unpleasant)
→ Examples: Extreme cold; Electric shock
10
Q
What is a motivating operation? What are its types?
A
- Any procedure that affects the appetitiveness or aversiveness of a stimulus
-
Deprivation: prolonged absence of an event increases appetitiveness
→ p.ex: going without food for a long period of time obviously increases the appetitiveness of food, thereby increase its ability to serve as a reinforcer for certain types of behaviour -
Satiation: prolonged exposure to an event decreases appetitiveness
→ food is much less effective as a reinforcer if a rat has just eaten a large meal
→ p.ex: being forced to eat a 9th slice of pizza
11
Q
How do we measure a behaviour?
A
Rate, duration, speed, latency, errors, reliability
- The definition must refer to some observable aspect of the individual’s behavior and be clearly defined
12
Q
What are some aspects of measurements of behaviour?
A
- Measures should be objective, and unambiguous (clearly defined)
- We often use more than one measure to assess behaviour in an experiment
- Ensures they can be replicated across time and place
→ this notion is the most important - Reliable measures have high inter-rater reliability
13
Q
What is inter-rater reliability?
A
- Measures/ratings should be reliable across multiple observers
- Ideally, behavior is measured by 2+ observers
→ Sometimes only a subset of the behavior
→ most important for qualitative research to ensure they reach the same scores - What is an acceptable IRR rate? 75%
14
Q
Explain rate of response as a measure.
A
- One of the more popular measures
- The frequency with which a response occurs in a certain period of time
- This method of measurement is most appropriate when the response is of brief duration, with a well-defined start and finish
→ it’s a very easy way to quantify something
→ Example: number of phone pick-ups per hour; the # of cigarettes smoked in a day
15
Q
Explain duration as a measure.
A
- The length of time that an individual repeatedly or continuously performs a certain behaviour
- This measure is appropriate when either increasing or decreasing the length of time the behaviour occurs
→ a student will want to increase time spent studying while decreasing time spent watching TV
→ Example: Total time spent on your phone each day; total time spent studying - Why is this not an ideal measure? It doesn’t take into account the quality of time spent studying
16
Q
Explain speed as a measure.
A
- How quickly or slowly a behaviour occurs, or the rapidity with which one progresses through some type of distance
- Very popular in aging research and used to measure processing speed
→ Example:
→ Length of time for a rat to run through a maze; Length of time for you to finish reading this lecture
17
Q
Explain latency as a measure.
A
- Length of time required for the behaviour to begin
→ with respect to classical conditioning, the strength of conditioning can be measured not by amount of saliva, but the length of time it takes for the dog to start salivating - Examples:
→ How quickly the rat begins running once it’s placed in the maze
→ How long you waited to start reading Chapter 2 since it was assigned
18
Q
Explain number of errors as a measure.
A
- Measures any behaviour in which responses can be categorized as right or wrong
- Examples:
→ the number of errors a student makes on an exam
→ the number of wrong turns a rat takes before it finishes all the food
19
Q
Name the types of research designs.
A
- Descriptive research, experimental research, group design, single-subject design, use of animal research
20
Q
To measure learning is to measure ___ in behaviour
A
changes
21
Q
What is descriptive research? What are its types?
A
- Naturalistic observation, case studies
- Involves simply describing the behaviour and the situation within which it occurs
→ p.ex: describing a group by obtaining information from participants through interviews - This research does not involve the manipulation of any variables
-
Naturalistic observation involves the systematic observation and recording of behaviour in its natural environment
→ Such behaviours may not occur if removed from environment -
The case study approach involves the intensive examination of one or a few individuals
→ These can be done in natural settings or detailed examinations in more structured settings (like in a clinician’s office)
22
Q
What are some major problems with naturalistic observation?
A
- Difficult to assess cause and effect (or the functional relationship)
→ which variables are important to determine the observed behaviour?
→ p.ex: if you study childhood aggression by observing children interacting on a playground, you may see varying sets of behavior of aggression, but it’s difficult to determine why these behaviors are occurring because researchers cannot intervene or interact with the participants - Can’t control other variables
→ this method is often insufficient for gaining a full understanding of a behavior and variables that influence it
→ the only way to determine that is to perform an experiment - Behavioural scientists often have many preconceptions about behaviour because it is their job to study behaviour and they can be susceptible to biases
→ to avoid biases, researchers attempt to define their variables objectively and make observations in a consistent and unified manner
23
Q
What are some major problems with case studies?
A
- takes a good deal of time, thus generalizations are based on very few cases (not representative of a larger group)
- Cannot answer certain questions about certain behaviour
–> p.ex: we can’t say that when a person falls off a ladder, they develop a fear of heights just because it occurred in one specific case - Difficult to establish causation; because case studies observe one person, we have no way of knowing if the observation is the norm or the exception
24
Q
What is experimental research? What are its types?
A
- Control group designs, single-subject designs
- Experiments to discover cause-and-effect relationships between environmental events and behavior
→ p.ex: do children who like to read to better in school, or do brighter kids simply like to read more than other children - One or more independent variables are systematically varied (manipulated) to determine their effect on a dependent variable
25
What is a control group design?
- Type of experimental research
- Individuals are *randomly assigned* to either an experimental group or a control group
→ **Experimental/treatment group**: Individuals are exposed to a certain manipulation or treatment (i.e. food deprivation)
→ **Control group**: individuals are not exposed to the manipulation (i.e. no food deprivation)
- Control groups are useful because they...
→ Allow us to establish a baseline to which we can compare it to
→ p.ex: this product is good because when used, there is an improvement compared to this baseline (control group)
→ Take into account *confounding variables*
→ Allows us to see if there is a *significant difference* between group
26
What are the benefits of control group design?
- Can determine cause-and-effects (functional relationship)
- Random assignment ensures various characteristics of the subjects in each group are likely to be evenly distributed
→ the demographics should be the same between the groups
→ sex, age and education are variables that need to be considered when we have a control group, even with random assignment, otherwise the effect may be confounded by these variables
27
What are limitations of control group designs?
- Requires large number of subjects (typically more than 10)
→ the larger the # fo subjects, the more trustworthy the results
- Ignores individual difference
- Focus on the *average* performance of all subjects in each group
→ little attention is given to the performance of individual subjects, even if some differ markedly from the average
- Results are often analyzed and interpreted only at the end of the experiment, rather than during
→ in some situations, this may be very undesirable
→ p.ex: looking at a drug over an 8 week period, we can assess every week because it will likely start having its effect sooner than the 8 weeks, this would allow us to adjust as we go
28
What is the difference between Between-subjects design and Within-subjects design?
Between: The control and treatment groups undergo different treatments
Within: The control and treatment groups undergo the same treatments
29
What is a single subject design? What are its types?
- Type of experimental research
- Require only one or a few subjects to conduct an entire experiment
- Problem of generalizability of the results to a broader population
1) **Single comparison design (AB)**
- behaviour in a baseline condition is compared to behaviour in a control condition
- Days 0 - 7 in phase A of baseline, Days 8 - 14 in phase B of treatment
2) **Reversal design (ABAB)**
- Repeated alternations between baseline and treatment
--> Days 0 - 7 in baseline, days 8 - 14 in treatment, days 15 - 21 in baseline...
- If the behaviour systematically changes each time the treatment is instituted and later withdrawn, then a functional relationship (cause-and-effect) has been demonstrated
3) **2-treatment reversal design (ABCAC)**
- When the 1st treatment isn’t proving to be effective enough, we increase the punishment (from B treatment to C treatment)
→ The main disadvantage of this is that the behaviour must revert to its original baseline frequency when treatment is withdrawn, otherwise we can’t know if the treatment had any effect
30
What do animals have to do with understanding human behaviour?
- **Genetic level**: common molecular mechanisms
- **Behavioural level**: common principles of learning
31
What are the advantages of using animals in research?
- Can control and manipulate their genetic makeup
→ genetic knock-out models
→ transgenic models
→ conditional gene expression techniques:
→ turn genes or cells ‘on’ or ‘off’; drugs; virus
- Can control their learning history
- Can control their schedule
- Can control their experimental environment (have them be raised in identical environments)
- Some research cannot ethically be conducted with humans (i.e. drug tolerance studies)
→ p.ex: drug to cure cancer can’t be tested off the bat with humans
32
What are the arguments against using animals in research?
- The findings from animal research have *limited applicability* to humans
→ Physiology, cognitive ability
→ Difficulties translating treatments from rodents to humans (a lot of the side effects seen in rodent testing will translate poorly in humans)
- The ethics of animal testing is controversial
→ Animal care committees to ensure humane conduct of all research with animals
33
