When the influence of an extraneous variable is different for the various groups.

If groups are equivalent on every variable except for one, then that one variable is the cause of the difference between the groups.

Control technique that equates groups of participants by ensuring every member an equal chance of being assigned to any group.

* one of the best techniques used

Randomly assigning a sample of individuals to a specific number of comparison groups.

*a control technique to equate groups of participants.

 - accomplished by ensuring that every member has an
   equal chance of be assigned to any group.
*provides maximum insurance that groups are equal

*Eliminates systematic differences between groups
*Doesn't eliminate extraneous variables, but randomly distributes them across groups.

*The MOST AND IMPORTANT & POWERFUL CONTROL TECHNIQUE

refers to the statistical characteristic of equiprobability (equal in probability) of events.

Using any of a variety of techniques to equate participants in the treatment groups on specific variables.

 - should be done with variables that differentially affect
   the DV in the various groups

- e.g. intelligence, age, gender

The extraneous variable used in matching.

A second means of controlling the extraneous variables through matching is to build the extraneous variable into the research design.

- make the extraneous variable another IV in the study

- should be used only when you are interested in the effect of the extraneous variable.

Control of measured extraneous variables during data analysis. Statistical control is much more important in quasi-experiment designs than in randomized experimental design because quasi-experimental designs lack random assignment.

A matching technique that matches participants on the basis of the temporal sequence of administering an event.

Example:

-research question-a researcher wants to know the effect of students   having the freedom to choose snack breaks on their classroom productivity.

-each control participant is "yoked" to an experimental participant

-experimental participant takes a self-chosen snack break

-the control "yoked: participant is given a snack break by the teacher.

You must treat the different groups in the same way during the conduct of the experiment, except for administration of the different levels of the independent variable.

Counterbalancing

                           Treatment Conditions

                     A                                             B

                    S1                                           S1

                    S2                                           S2     

                    S3   ←   Same Participants → S3

                    S4                                           S4

                  S5                                          S5

A technique used to control for sequencing effects. 

           (order effects, carryover effects)

*used only with repeated measures (within participants)
 designs.

*All participants receive all treatment conditions.

*All participants receive all levels of at least one independent variable.

A sequencing effect arising from the order in which the treatment conditions are administered to participants.

e.g. practice, fatigue

1. Order Effect

2. Carryover Effect

A sequencing effect that occurs when performance in one treatment condition affects performance in another treatment condition.

e.g. effects of treatment A still affecting participant when engaged in treatment B.

Sequence Order is randomly determined for each individual.

*in effect the experiment is replicated for each participant using different counterbalancing sequences. When you have three treatment groups and three treatments, there are six possible sequences in which the conditions can be presented to the participants: [1,2,3] [1,3,2] [2,3,1] [2,3,2] [3,1,2] and [3,2,1]. *you do not decide the sequence; you must use a random process such as a table of random numbers or a random number generator.

Administrating the treatment conditions to each individual participant in more than one order.

* may not be feasible with long treatment sequences.

Example:

-typically done with two treatments

-take treatment in one order and then reversed order.

*each participant takes the treatment conditions first in one order than in the reverse order. ex:2 Pepsi challenge. 2 colas Pepsi/Coke. 1 assessment of whether participants like cola A (Pepsi) or cola B (Coke) in AB order. Participants would then taste the cola B (Coke) and then cola A (Pepsi) then make an assessment. Marking the sequence ABBA. *Many observed difference between the averages would not be attributable to carryover effects or order effects because they would have been equalized, or held constant, across conditions.

Administering different sequences to different groups of participants.

Enumerating all possible sequences and requiring different groups of participants to take each of the sequences.

*all possible sequences of treatment conditions are used

*participants randomly assigned to sequence

*N! = N multiplied by each number below it

-2!=2x1=2

   * AB, BA

-3!=3x2x1=6

  * ABC,ACB,BAC,BCA,CAB,CBA

-4!=4x3x2x1=24

**rarely used with more than 3 conditions because number of possible sequences (N!) is too large.

Limitation of complete counterbalancing is that when the number of treatment conditions is large, the number of possible sequences becomes unwieldy.

You can determine the number of possible sequences by calculating N! (called N factorial) The number of treatment conditions is symbolized by "N", and the "!" symbol means you multiply N by the numbers falling below it: N times N-1, N-2, and so on until you reach one. For example if you have three groups; then N is 3 and N! is 3 times 2 times 1 (which is 6). For four groups: 4 times 3 times 2 times 1 (which is 24)...

* Most commonly used technique

* not all possible sequences are used

*Criteria- for sequences enumerated

1) each treatment condition must appear an equal number of times in each ordinal position.

2) each treatment condition must precede and be followed by every other condition and equal number of times

*sequences determined by formula  -1,2,n,3,(n-1),4,(n-2),5

In the case of a caffeine study with four treatment conditions, the first sequence would be ABDC, or 1,2,4,3. If an experiment consisted of six treatment conditions, the first sequence would be ABFCED, or 1,2,6,3,5,4. The remaining sequences of incomplete counter balancing technique are then established by incrementing each value in the preceding sequence by 1. For example in the caffeine study, in which the first sequence is ABDC, the second sequence is BCAD. Naturally, to increment the last treatment condition, D, by 1, you do not proceed to E but go back to A.

A treatment condition affects' participants performance in a later condition in one way and in another way when followed by a different condition.

ex: the carryover from treatment A is 4 units when it is immediately followed by treatment B, but the carryover from treatment A is 2 units when it is immediately followed by treatment C.

-if suspected, don't use repeated measures.

Neither the experimenter nor the research participant is aware of the treatment condition administered to the participant.

Giving the participant a bogus rationale for the experiment.

-omission of or altering the truth of information given to
 the participant during a research study.

-used when there is no other way to gain the knowledge
 and risk does not outweigh the benefit of the information.

-must keep the false information constant for all
 participants.

An oral report in which the participant retrospectively recalls aspects of the experiment.

An interview of the participant after the experiment is over.

A participant's oral report of the experiment, which is obtained as the experiment is being performed.

Groups of participants who are stopped and interviewed at different stages of the experiment.

Obtaining a participant's perceptions of the experiment after the completion of each trial.

A method that requires participants to verbalize their thoughts as they are performing the experiment.

The biasing influence that can be exerted by the experimenter.

*Control of Recording Errors

*Control of Attribute Errors

*Control of Experimenter expectancies

*Increase number of experimenters

*Observe behavior of experimenters

- make researchers aware of the necessity of making careful
  observations. (may require training before data collection).
- Use multiple data recorders (e.g. video cameras, computers) or
  observers.
- Have participants make reaponses on a computer.

         Control of attribute errors:

-referring to the attributes of the experimenter.

-use the same experimenter in all treatment conditions unless the
  treatment condition interacts with attributes.

The influence of experimenter attributes should not significantly affect the mean differences among treatment groups. *The key point is to never use one experimenter in one condition and a different experimenter in the other condition; this would make the groups different not just in the independent variable but also on the experimenter.

1. Blind technique

2. Partial Blind Technique

3. Automation

A method whereby knowledge of each research participant's treatment condition is kept from the experimenter.

e.g. double blind procedure

A method whereby knowledge of each research participant's treatment condition is kept from the experimenter through as many stages of the experiment as possible.

The technique of totally automating the experimental procedures so that no experimenter-participant interaction is required.

If you could use only one control technique, which one should you use?

Matching by including the extraneous variable into the design of the study

Assume that you wanted to investigate the effect of caffeine on a person's ability to identify the number of times the letter q appeared on a page filled with a random list of letters. To control for the effect of a person's reaction time, you divided the people into those that had high and low reaction times and then included this difference in reaction time as another independent variable in the design of your study. By controlling for the possible influence of reaction time in this way you used?

If you identify extraneous variables that are correlated with the dependent variable and control for them by matching participants on an individual basis and then randomly assign the matched participants to groups, which control technique have you used?

Assume you wanted to find out if alcohol increased a person's aggressiveness. To test this hypothesis, you wanted to test people's aggressiveness while they were under the influence of alcohol and while they were not. However you know that you are asking people to perform under two conditions and that performing once might change their performance on the second occasion. To control for this effect you elect to?

Automate the experimental procedure so that the experimenter does not interact with the research participant

If you wanted to control for the expectancies that the experimenter might have regarding the outcome of the experiment, you might

The outline, plan, or strategy used to investigate the research problem.

Designs that do not control for many extraneous variables and provide weak evidence of cause and effect.

Only use when a stronger design cannot be used to answer the research question.

1) One Group posttest-only design

2) One Group pretest-posttest design

3) Posttest-only design with non-equivalent groups

Administration of a posttest to a single group of participants after they have been given an experimental treatment condition.

-rarely useful because no pretest or control group

-almost all threats to internal validity apply

-is useful only when specific back ground information exists on the DV.

Treatment                                    Posttest Measure

     X                                                         O

Design in which a treatment condition is interjected between a pretest and posttest of the dependent variable.

-most threats to internal validity exists

-to infer causality must identify and demonstrate that internal validity threats do not exist.

Ex:

O = students' reading      X= new reading curriculum    O = students' reading                                    

measured at beginning    new reading curriculum          measured at the end of

school year                     employed throughout            school year.

                                     school year

Threats to internal validity not controlled for: history,testing,regression,

instrumentation, and maturation.    

                                                                      Pretest Measure     Treatment      Posttest Measure    

                                                                               O                        X                       O

                                            \        Compare     /                            

Design in which the performance of an experimental group is compared with that of a nonequivalent control group at the posttest.

                                Treatment       Posttest Measure

Experimental Group          X                      O

                                                                   >  Compare

Control Group                                           O

-Experimental Group= School A adopts new reading curriculum (X)

-Control Group= School B uses old reading curriculum

-Both schools postest reading at the end of the year.

1. One-Group Posttest-Only Design

2. One-Group Pretest-Posttest Design

3. Posttest Only Design with Nonequivalent Groups

Experimental design with random assignment to experimental and control groups.

The group of participants that do not receive the active treatment condition and serves as a standard of comparison for determining whether the treatment condition produced any causal effect.

The group of participants that receives the treatment condition that is intended to produce an effect.

What the experimental group participants' responses would have been if they had not received the treatment.

Designs that effectively control extraneous variables and provide strong evidence of cause and effect.

Improved inernal validity achieved by eliminating rival hypotheses.

-with control techniques

-with a control group

   group that does not get the IV or gets some standard value

   serves as source of comparison to experimental group

   controls for rival hypothesis

Groups are produced by random assignment, and the different groups are exposed to the different levels of the independent variable (uses a between subjects IV)

*Posttest-Only control-group design

-random assignment to groups creates equivalence

-use of control group eliminates most threats to internal
 validity

Weaknesses of design

-does not guarantee equivilance of groups- particularly with small sample size (N<30).

Between-participants designs in which participants are randomly assigned to groups.

Administration of a posttest to two or more randomly assigned groups of participants that receive the different levels of independent variable.

                                                  Treatment    Posttest Measure

                         Experimental             X                      O

                   /          Group

Sample of

Research                    Randomly

Participants                assigned to

                    \

                         Control Group                                     O

-participants randomly assigned to groups/ -IV = Social Skills Training

Treatment group - X rcvd training / Control Group - did nto rcv training

-DV= posttest measure (O) for both groups

All participants receive all conditions (uses a within-subjects IV)

Another common name for a within-participants design.

All participants receive all conditions, and a posttest is administered after each condition is administered.

Participants included in all conditions (also known as repeated measures designs).
Counterbalancing necessary to eliminate linear sequencing effects
Within-participants receive all conditions, and a posttest is administered after each condition is administered.

Ex: participants=elementary school children

-IV= breakfast type

  * cereal, oatmeal, no breakfast

  * each type on three different days

-DV= cognitive performance

  * series of cognitive tests after breakfast

Strength: increased sensitivity because effects of individual differences are controlled & fewer research participants required

Weaknesses: sifficult for participants & potential problem of differential carryover effects.

Contains both between participants and within participants variables.

Pretest-Postest Control Group Design

Advantages of including a pretest:

  *can assess the effects of readomization

  *determine if ceiling or floor effect has occurred

  *allows use of analysis of covariance to statistically
   *control for pretest differences

  *allows researcher to assess the change in
    dependent variable from pretest to posttest.

Disadvantage of including a pretest:   * may not generalize to situations with no pretest.

Type of IV where different participants receive different levels of the IV.

Administration of a posttest to two or more randomly assigned groups of participants after the groups have been pretested and administered the different levels of the IV.

                                                      Pretest Measure  Treatment    Posttest Measure

                   /     Control Group                   O                                           O

Sample of

Research               Randomly

Participants          assigned to

                    \

                         Experimental Grp              O                   X                      O

Pretest for social anxiety/Random assignment of 100 participants with social anxiety/ IV = treatment for social anxiety experimental group=anxiety reduction (N=50) /control group= no anxiety reduction treatment (N=50)

DV= level of social anxiety (posttest) /Results

Situation where participants' pretest scores on the DV are too high to allow for additional increases.

Situation where participants' pretest scores on the DV are too low to allow for additional decreases.

A statistical procedure in which group means are compared after adjusting for pretest differences.

Two or more IVs are studied to determine their separate and joint effects on the dependent variable.

They can include:

-only between participants variables

-only within participants variables

-both between and within participants variables (mixed)

Win lottery someone ask you for help

Identity theft, lose everything & someone asks you for help

Type of IV where different participants receive different levels of the IV.

Combination of levels of two or more IVs

e.g. person not getting adequate sleep and having
moderate level of caffine

The average score of the participants in a single cell.

The average score of all participants receiving one level of an IV.

The influence of one IV on the DV.

When the effect of one IV depends on the level of another IV.

When displayed graphically, an interaction yields non-parallel lines.

The effect of one IV on the DV varies with the different levels of the other independent variable.

Type of IV where all participants receive all levels of the IV.

A factorial design that uses a combination of within-participants and between-participants IVs.

A two-way interaction that changes at the different levels of the third IV.

The one-group posttest-only design, the one-group pretest-posttest design, and the posttest-only design with non-equivalent groups have ____ in common?

1. To control for some rival hypotheses

2. To serve as a comparison

The primary difference between a between-participants and a within-participants design is

If I have studied the effect that three dosages-15,30, and 60mg-of the drug Cymbalta has on depressed people and people with an eating disorder and have found that the lower dosages are more effective with people with eating disorders and the higher dosages are most effective with people with depression, I have identified

Factorial design based on a mixed model

If I have conducted an experiment that requires me to randomly assign 30 participants to two levels of one independent variable (15 in one condition and 15 in the other condition) and all 30 participants take all three levels of a second independent variable, I have used what type of design?

The probability of correctly rejecting the null hypothesis

Power of at least .80 is desired

 - i.e. we want to correctly reject a false null 80% of the time.

Factors that influence XXXXX

 - sample size

 - alpha level
 - effect size

   *the magnitude of the relationship between two variables in a population.

   *can be estimated at small, medium, and large

   *examine literature for approximate effect size estimation.

Should conduct XXXXX Analysis to estimate sample size.

The magnitude of the relationship between two variables in a population.

-An experiment that is conducted on a few participants prior to the actual collection of data.

-Serves several purposes:

 *establishes clarity of instructions

 *provides evidence that IV is being manipulated adequately

 *familiarizes researcher with the procedure

-Internet Studies - also complete it yourself

 *Give you information on how it feels to be in this type of study

 *Give information on if the study works in your browser and if the data is
   returned in the desired way

The type of organism that should be used in research studies..

Sample size should be determined by a combination of which of the following factors?

Behavioral Research Methods, Instruments, and Computers

Which of the following journals will be helpful in identifying a specific piece of apparatus or computer program to assist in data collection?

Conducted a pilot study

If you have pretested your entire procedure on a few participants prior to actually collecting data, you have 

1.Participant satisfaction from contributing to science.

2. Ethical function.

3. Educational function.

4. Methodological function.

Goal of Experimentation

Identify the casual effect of the IV on the DV
  - must have internal validity to do this
  - requires control of extraneous variables

* Type of errors
   - Random errors
   - Systematic errors

-eliminate differential influence - confounds

*Variables need to control

   1- Individual difference - e.g. maturation

   2- Outside Influence - e.g. history

   3- Experimental contamination - e.g. systematic errors

*Ways of achieveing control

   - Design of the experiment

   - Statistical adjustments

   - Incorporate control techniques into the research
     design.

1. Controls for the variables on which participants are matched.

2. Increases the sensitivity of the experiment

* Increases chances for locating an effect if one appears.

Chance probability 5 in 10 or less

-Individual matching

 *match case by case then randomly assign to groups

  *advantage
   -groups equated on potential extraneous variable

  *disadvantage

   -identifying the variables on which to match

   -matching participants increases as the number of 
    variables on which to match increases

   -some variables difficult to match
   -decrease in generalizability of results

randomized counterbalancing

intrasubject counterbalancing

complete counterbalancing

incomplete counterbalancing

Incomplete Counterbalancing

Jordan planned to have subjects in an experiment participate in three experimental treatment conditions, A,B,C, & D, one right after another. In doing so Jordan realized that he could have sequencing effects confounding his study so he had one group of subjects take the conditions in the order of A,B,D,C, another group take the condition in the order of B,C,A,D, a third group take the conditions in the order of C,D,B,A, and the fourth group take the conditions in the order of D,C,A,B. What counterbalancing technique did he use?

Retrospect verbal reports 

If I wanted to get some assurance that subjects were interpreting the experiment they participated in the same way, one method of doing this would be to interview the subjects right after they completed the study to find out what they thought of the experiment, what they expected to find, etc. This would involve?

*technniques used to gain insight into participants' perseptions of 
  experiment.

*retrospective verbal report: an oral report in which the participant
  retrospectively recalls aspects of the experiment.

questions to ask:

-What did the participant think the experiment was about?

-What did he/she think the experimenter expected to find?

-What type of response did the participant attempt to give & why?

-How does the participant think others will respond in this situation?

*Concurrent verbal reports: sacrifice groups, concurrent probing, think-aloud technique.

Attribute errors

If in conducting an experiment I wanted to test over 100 subjects and in doing so I had to make use of several experimenters to get the data needed within one semester. To do this I enlisted the aid of 5 graduate students, a white male, a white female, a Hispanic female, a black female, and a black male. I had each of these five experimenters collect data from five individuals in each of the four experimental conditions. In doing this I would be attempting to control for?

Example:

  gender and conformity study
  -using only females.

Disadvantages

- restricts the population size

-restricts generalization to the type of participants

 in the study

                                                                            Treatment         Posttest Measure

                                Control Group                                                         O

Sample of        Ramdomly assigned to
Research     

Participants         

                              Experimental Grp 1                       X1                       O

                              Experimental Grp 2                       X2                       O

The groups are randomly assigned to treatment conditions

The posttest-only design with nonequivalent groups and the pretest-posttest control group design are similar in that they both have a control and experimental group. However, the pretest-posttest control group design is a stronger design because?

                                                Caffiene Consumption

                                                                                Low        Medium        High

                       Not Deprived                                            Low        Medium        High

                                                                                        not            not             not

Sleep                                                    deprived     deprived      deprived

Deprivation                     

                      Deprived                                                   Low        Medium        High

                                                                                    deprived     deprived      deprived

* 2x2 design

   - number if numerials = number of IVs = 2

   - each number indicates the number of levels for each IV

   * IV1 = 2 levels

   * IV2 = 2 levels

*2x3 design

   - 2 IVs

   * IV1 = 2 levels

   * IV2 = 3 levels

Different main effects

 e.g. gender, attractiveness

In a study investigating how liking was affected by both makes and females who were either very attractice, somewhat attractive, or unattractive, Josh randomly assigned subjects to each of the six combination of conditions in the study. In this study the attractiveness and the gender variables would respresent?

Main effect of gender and the interaction

In the attraction and liking study previously mentioned, it was found that overall females were liked more than males. However, although attractive females were liked more than attractive males, unattractive females were liked less than attractive males, unattractive females were liked less than unattractive males. In this study a significant effect exists for the?

Factorial design based on a mixed model

Jordan wants to determine which of the two types of theraphy was most effective in ameliorating the effect of PTSD on vets returning from a war zone. He was aware of a trend to have returning soliders get involved in community projects and that there were reports that this was very effective in helping recovery and he wanted to compare the effectiveness of this method with the traditional psychotherapeutic method of treating PTSD. To conduct the study, he randomly assigned a group of returning vets with PTSD to community service or to get psychotherapy and then tested their level of PTSD every week for six weeks. The result was that the vets showed more rapid improvement in the community service group at every week of testing versus the standard psychotherapy group. This study used what design?

Strengths:

-more than one IV allows for more precise hypotheses

-control of the extraneous variables by including as an IV

-ability to determine the interactive effect of two or more IVs.

Weaknesses:

-using more than two IVs may be logically cumbersome

  - 2x2 design= 4 cells, 2 main effects, 1 interaction

  - 3x3 design= 6 cells, 2 main effects, 1 interaction

  - 4x4 design= 12 cells, 3 main effects, 4 interactions

-higher-order interactions are difficult to interpret.

- Examination of prior research literature can guide choice of design

-Many factors to consider

 - use of a control group

 - number of comparison groups

 - pretest(s)

 - within-participants or between-participants

 -number of independent and dependent variables

Institutional Animal Care and Use Committee

(IACUC)

-reviews research protocols for studies using nonhuman animals

-reviews research protocols for studies using human participants

-primary concern is participant welfare

-chosen participants depends on research question- but often precedence is used

-obtaining animals

  * albino variant of brown rat is most commonly used

  * the Animal Welfare Act regulates care and housing

  *the Guide for the Care and Use of Laboratory Animals

    (1996) provides a guide to using animals 
    appropriately.

-obtaining human participants

  recruitment depends on type of participant needed

  convenience sample of college students often used in
  psychological research

 -may not be representative of target population

*Obtaining human participants

  -using the Internet to recruit research participants is 
   becoming more common.

   *advantages: providing access to a larger and more 
    diverse sample than otherwise possible.

   *disadvantages: select groups may not have access to
     Internet & sample may not be representative of
     target population.

No matter how participants are recruited
 -important to report how participants were selected and
  assigned to research conditions (methods section)

How many research participants should be included in the research?

practicality must be balanced with the increased power that accompanies a large sample.

A. Power

B. Effect Size

C. Alpha Level

D. Significance level required

If I have identified the size of the relationship between the IV and DV, I have identified?

How will IV be manipulated and how will DV be measured?

Depends on nature of research

  - may involve active participation by researcher

  - a particular type of apparatus

    ex: computers, MRI, EEG

Behavioral Research Methods is a good source to identify research instruments.

-detailed description of how experiment will be conducted.

-important to allow for future replication

Involves specification of the sequence of events

-Animals-specifying everything from care in the lab to
 performance in the study

-Humans-specifying everything from scheduling of
 participants to debriefing

-Internet study- have a series of "warm up" trials

Humans - availability of the researcher and participants

Rats - leave the light on 24/7 so they will not form a schedule based on the light/dark phase of the lighting schedule.

-Informed consent necessary unless waived by IRB

-Consent form should include the following elements:

*basic information about the study

*details about procedure of study including possible risks

*any potential benefits that might be derived

*the type of compensation provided and compensation penalty for early withdrawl

*for sensitive topics(e.g. depression, abuse) extra information for individuals who need assistance

*participants must be informed that they can withdrawl from the study at any time without penalty

 - if using questionaire, a statement indicating that participant can refuse to answer any question
    without penalty.

*participants must be informed as to how the records and data obtained will be kept confidential.

-should be precise, but not overly complex

-refrain from certain phrases

 * pay attention, relax, and ignore distractions

 *Requesting incompatible tasks performs at same time.

-language should be simple and on the level of the
 participant --clear, unambiguous, and specific

-warm-up trials can assess understanding of XXXX

-follow design

Functions

-ethical

 *undo any deception involved in the study

 *decreases stress or risk if possible

 *attempt to return participants to pre-experimental state

-educational

 *this mostly applies to students participating in research for class credit

-methodological

 *to establish effectiveness of IV or deception

 *to judge accuracy of participants suspicions or perceptions

-sense of satisfaction
 * participant feeling as if theycontributed to science and society

-face to face generally preferred

-begin by asking if participant has questions

-question participant to determine if all aspects of the study were clear

-if deception was used

 *attempt to determine if participant discerned true nature of study

 *explain the purpose of using deception

-ask participant not to reveal details of experiment to other participants.

- To test the effectiveness of the IV

- To gain experience with the procedure involved in data 
  collection

- To test the effectiveness of the DV