PSYC 3000. C1-1 Know the five steps involved in the scientific method. “Know” means to be able to list and briefly describe each step (14-18). –ID problem:

PSYC 3000

PSYC 3000C1-1 Know the five steps involved in the scientific method. “Know” means to be able to list

and briefly describe each step (14-18).

– ID problem: form hypothesis (a stated relationship between 2 or more variables)

– Design experiment: control for extraneous variables

– Conduct experiment– Analyze, interpret data– Communicate results

Advantage of scientific approach…

C1-2. What is the advantage that the scientific method offers over the other five non-scientific methods (19,3-20,0)?

Objective observation: no personal bias/opinion

Characteristics of science….

C1-3. List and describe, and give the relevance of each of the characteristics of the scientific approach (control, operational definition, replication). Be prepared to provide an operational definition of some concept that I might give you on the test (e.g., intelligence, anger).

– Control: eliminating influence of extraneous variables

– Operational define: terms and variables defined by steps or operations used to measure them

– Replicate: data must be reliable/replicable

*Operationally define “Intelligent”

Operations Def…

An operational definition of “Stupid” might be

1. Dumb

2. Unable to finish (no longer scrubbing) washing one’s hair within 5 minutes

3. Acts weird in public

4. Burps and acts like a wise-guy when sitting in a classroom

5. Doesn’t think well

Four objectives of Science

C1-4. Know the four objectives of science. On an exam, I might give you an

example and request that you use the example to illustrate each of the objectives (26-29).

– Describe: operationalize/define carefully– Explain: determine cause– Prediction: determine when event will

occur– Control: produce event by manipulating

antecedents (“make behavior happen”)

Five Assumptions ….

C1-5. Your author identifies five assumptions underlying science, starting with “Determinism”, with each point identified by a subheading. Be able to list and briefly describe each of these points (29-31).

Determinism: causes of behavior and they are accessible (all behavior is caused)

Reality in nature: events/objects in nature are real (vs. Plato: only real through perception)

Rationality: events occur for logical reasons (can use reason to understand)

Regularity: same laws of nature apply everywhere/all the time

Discoverability: it’s possible to discover regularities and causes

• Why don't pigs drive cars?

• Why don't pigs drive cars?

• They would become road hogs!

Quiz!

Reason is able to be used to help us understand the world, as the world operates in a logical manner. This is an example of:

A. Determinism

B. Reality in nature

C. Rationality

D. Regularity

E. Discoverability

Descriptive Research Approaches…. (chapter 2!)

C2-1. Know the primary characteristic of the descriptive research approach (45,5) and for each of the descriptive research methods I cover in class. Be able to briefly describe what each is, and list the advantages and disadvantages (46-72).

Descriptive Research Approach: observational, non-scientific– Accurately describes events/situations– No attempt @ discovering cause and effect relations

Several methods (a sample):Secondary records: birth records, census, video recordings, etc.

ADV: no reactivityDisADV:

Selective Deposit—only some events recordedSelective Survival—only some records survive

Example: Crack babies—reported in 1990s and beyond, but not available pre-1980s

Naturalistic Study: observe naturally occurring behavior ADV: little reactivity; no artificiality

DisADV: causes not discovered; time-consumingExample: Jane Goodall & chimpanzees; Diane Fossi & apes

Descriptive Research Approaches….

Correlation Study: measure 2 or more variables & determine degree relationship between them.ADV: able to predictDisADV: third variable problem Example: SAT and GPA, Height & Weight

Case Study: observe individual, event, or groupADV: intense observation of usually rare eventDisADV: little control; can’t generalize; can’t ID causesExample: Freud’s methods

Longitudinal Study: study individuals or some variable over a relatively long time periodADV: can see developmental changes over timeDisADV: time consuming; no cause and effect; lose participantsExample: follow set of students from childhood to adulthood (developmental psych.)


Cross-sectional: study different age individuals on some variable. Similar to longitudinal, but don’t follow over time—instead, get different aged participants.

ADV: can examine developmental changes or skills at various age levels; less time-consuming

DisADV: no cause and effect; cohort effect—different age groups may have been exposed to events that changed them (like 9/11 - a confound)Example: study participants aged 2, 4, 8, & 10 years

Survey: snapshot of current attitudes, beliefs, etc. (“verbal reports”)ADV: may be predictive; some insight into current eventsDisADV: easily biased; people are poor observers of own behavior (inaccurate); positive self-presentation by participantsExample: Gallup polls, Nielsen’s, etc.


C2-2 Know my point to be made in class regarding the survey as a verbal report versus direct observation of behavior. Also be able to recognize examples of and generate original examples of open-ended and close-ended survey questions. Be able to identify “double-barreled” questions as well as knowing the important points listed under the headings “ordering of the questions” and “questionnaire length” (65-66).

Verbal Report is influenced by many things – language is easy to emit – often inaccurate. Comparisons of direct observation vs. verbal self reports reveal self report inaccurate!

But… survey can give some preliminary information so….• Ordering of questions: demographic questions first, because

they are easy and “lead into” harder ones.

• Short is better; in person is better (mail return is <2%).

• Open-ended questions: Answer any way you want to.

• Close-ended question: Limiting responses; easy to score.

• Double-barreled: Two questions with only one response…how score??

• Biased: Slanted; leading to answer in a particular way.

Population, Sample, & Random Sampling

C2-3 Know the difference between the terms “population” and “sample” and know what “haphazard sampling” and “random sampling” are (67).

Population: All people you are interested in

Sample: a subset of the population you are interested in

Haphazard Sampling: nonprobability; obtain participants where you find them. Could be a biased sample.

Random Sampling: every member of population has an equal chance of being selected.

Random Sampling

1 = Random sampling, 2 = Haphazard sampling

You select every other person in the Turlock phone book for your study (your pop is all the folks in Turlock)

Random Sampling


You put all the names of everyone in the Turlock phone book into a hat, shake it up, then pull out 20 names that will be in your study.

Random Sampling


You put all the names of everyone in this class into a hat, shake it up, the take out 10 names who will be in your study. The class is your population of interest.

The Experimental approach

C 3-1 What occurs in an experiment?

An experiment involves the independent variable being manipulated and others being controlled (the potential extraneous variables)

Advantages of Experimental approach

C 3-2 Know the advantages of the experimental approach—three are listed. Also know the disadvantages (also three listed) (87-89)

Advantages of the experimental approach:* causal relationships can be stated with confidence* precisely manipulate one or more variables* usefulness—leads to solutions/more research

Disadvantages* poor generalization from lab to “real” life* experiments are time consuming

Field vs. Lab

C3-3 Compare and contrast field experimentation and laboratory experimentation. What are the advantages and disadvantages of each? (92-95)

Field ExperimentExperiment conducted in real life setting (need

manipulation)ADV: little artificiality (better generalization)DisADV: little control over extraneous variables (unsure of results due to IV or EX)

Lab ExperimentADV: excellent controlDisADV: Artificial (little generalization)

• Why do squirrels spend so much time in trees?

• Why do squirrels spend so much time in trees?

• To get away from the nuts on the ground!

Researchable or not?

C4-1 Be sure to identify and generate examples of ideas that are and are not researchable (108)

Researchable and non-researchable ideasa. Researchable: (must be observable /measurable/

testable – empirical)

b. Non-researchable: typically morality/religion/value judgment issues – non empirical!

A man’s soul weighs more than a woman’s soul

:450 of 35

0%0%

1. Researchable

2. Not researchable

The fastest way to Modesto is via the

back roads (Santa Fe) versus 99.

:450 of 35

0%0%

1. Researchable

2. Not Researchable

Praying to God will reduce the length of a cold

:450 of 35

0%0%

1. Researchable

2. Not researchable

How long one can hold a 10 lb weight over their head

:450 of 35

0%0%

1. Not researchable

2. Researchable

Why review the literature?

C4-2. What are the benefits of reviewing the literature? (109) I found two points.

a. find out what’s been done

b. point out methodological problems

Criteria for research problem

C4-3. Know the criteria in defining a research problem (“research question” is probably a better way to state this instead of conceptualizing all research as ‘problems’) and be able to generate examples or recognize examples of good and bad research questions (117-120).

Criteria for the selection of a research question

a. states relation between 2 or more variables (often specifies direction of relationship)

b. should be empirically testable

Scientific vs. Null hypothesis

C-4-4. Be able to clearly distinguish between a scientific hypothesis and a null hypothesis (120-122). Why is a null hypothesis tested and what is said when it is rejected? (122,0)

Hypothesis: A statement specifying a relationship between 2 or more variables and is testable

Scientific hypothesis: Predicted relationship among variables being investigated.

Null hypothesis: A statement specifying no relationship among variables being investigated.

IV has no effect of DV if null is correct.Reject the null = IV affected DV significantly

You state that caffeine will improve test scores

:450 of 35

0%0%

1. Scientific hypothesis

2. Null Hypothesis

You believe that caffeine will decrease test scores

:450 of 35

0%0%


2. Null Hypothesis

You state that you drinking watermelon juice prior to taking a test will not result in

any change in one’s test scores.

:450 of 35

0%0%


2. Null Hypothesis

All about ethics in research

C5-1 Ethics in research

- Milgram, Syphillis, huh?

- APA code for ethical treatment

- Types of stress: Physiological, psychological

- Exempt vs. nonexempt- Informed consent

- Some aspects: Confidentiality, special populations, right to withdraw, right to results, right to effective treatment,

- Overriding principle, benefit outweigh cost

IV and DV

C6-1.Be able to define, recognize examples of and generate examples of variable, IV, DV, discrete variables, continuous variables, qualitative variables, and quantitative variables (191-193).

Variable: measurable characteristic

Independent Variable: Antecedent variable/manipulated

Dependent Variable: Variable measured; detects influence of IV

Qualitative Variable: Vary in kind

Quantitative Variable: Vary in amount

The amount of caffeine in the bloodstream after drinking coffee is ….

:450 of 35

0% 0% 0%0%0%

1. Independent Variable

2. Dependent Variable

3. Factorial design

4. Anova

5. None of these

How fast one runs a mile after work is an example of

:450 of 35

0% 0% 0%0%0%



3. Factorial design

4. Anova

5. None of these

How fast one hands out a test (slow or fast)

:450 of 35

0% 0% 0%0%0%0%



3. Factorial design

4. Anova

5. None of these

6. Could be either

IV and DV

C6-2. Know the two requirements for a variable to qualify as an independent variable (194, 1). Be able to list and describe the three methods the author gives for manipulating variation (1194-197). Note that these categories are not independent of one another (lecture).

IV must:

be variable

be able to be manipulated

Three methods of manipulation:

• presence/absence

• amount of variable

• type of variable

IV and DV

C6-3. Be prepared to recognize and generate examples of operationally defined IVs (205-207).

Be able to give examples.

Give an operational definition of an IV and a DV.

E.g. provide an example of manipulating self esteem as and IV, or measuring self esteem as a DV.

Construct Validity

C6-4. Know the definition of "Construct Validity” and how it is established (2 methods listed in 209, 2-3). Also know the three methods for checking on the manipulation of the IV (210, 3-211,3).

Construct: a consistent set of behaviors with a label on it, i.e., friendly, athletic, etc.

Construct Validity of the IV (operationally defined): The extent to which an abstract construct can be inferred from the operational definition of that construct.

Establish by…• Clear operational definition of abstract construct• Showing convergent (expected outcomes) and

divergent data (diff measure and no relation to outcomes)

Manipulating the IV

C6-4. Know the definition of "Construct Validity” and how it is established (2 methods listed in 209, 2-3). Also know the three methods for checking on the manipulation of the IV (210, 3-211,3).

Checking on manipulation of IV:

Interview participants: to ensure that the IV had the desired effect on them.

Behavioral Indicator (blood pressure/EEG/GSR) (or physiological indicators)

Pretesting/pilot data

More than one IV

C6-4. What is the advantage of using more than one IV? Be able to define “Interaction” and give examples of it (212).

Number of IVs:More than 1 gives interaction data (in addition to main

effects) = Different effect a variable has on different levels of other variables

Test scores

Vitamins No vitaminsInteraction Graph

Make experiment only as complex as needed to show relationships clearly!

factorial design (2X2; 2X4; etc.)

Exercise

No Exercise

• How many 9's are there between 1 and 100?

• How many 9's are there between 1 and 100?

• 20

DV

C6-6. Know the definition of “DV”(213,1). What are the tasks of a DV (214,2-215,1). Your author lists three aspects of an experiment related to the DV that must be considered by the researcher. These include selecting the DV, the subject’s motivation level (“taking it seriously” p 215,1) and “cooperating” (215,1).

DV: The behavioral variable designed to measure the effect of the variation of the independent variable.

Dependent variable: should be:

• Sensitive

• Must ensure participant being serious and not “cooperative”

Reliability and Validity

C6-7. Know the difference between and the definition of reliability and validity. How are each established (217-221).

Reliability of DV:Extent to which the dependent variable is consistent or stable over

time. Measure must be reliable over time.Do replications and see if get similar measures on the DV

Validity: The extent to which the DV measures what you want to measure

Check by:Convergent data: extent to which similar measures correspond to

measure used

Divergent data: extent to which dissimilar measures do not correspond to measures used

Disguising the DV

C6-8. What are the benefits of disguising the DV (221,2)? (through distractions)

• Participants take measure seriously

• Participants’ demand characteristics controlled for

Internal Validity

C7-1. Know the definitions of internal validity and extraneous variable (229). What are the two methods of controlling extraneous variables (230-232). Know the point made in 232,2 regarding “The difficulty frequency lies in identifying those variables.”

Internal Validity:Extent to which one can accurately state that

the IV produced any observed effects on DV.

Extraneous variables controlled by:– Elimination– Keep constant across groups/eliminate

difference between groups (random assignment)

You can’t identify all “participant variables” (possibly confounds such as gender, age, IQ, experience, etc.) . Can only control through random assignment (by balancing experimental and control groups)

Extraneous variables

C7-2. Be able to list and define the six extraneous variables that Christensen notes need to be controlled for. Be able to give an example of each and to recognize examples of each (233-240). (Most relate to pre-post-test design)

History—something which occurs between a pre & post test and may effect DV

Maturation—biological/psychological changes that occur with time passage

Instrumentation—assessment tool/person changes during course of study. Change in DV may be due to this. (combat with interobserver agreement)

Statistical Regression—high and low scores in a distribution will tend to regress back toward the mean (change in DV may be due to this) (usually happens with tests to raise very low scores) prevent: don’t use highs and lows, use a range. detect: use control group which should mimic stat. reg. (byproduct of poor selection)

Selection—when the selection of participants is NQR (not quite right) and the various groups are unequal in composition. Thus characteristics of participants may cause observed effect.

Mortality—unequal loss of participants from the various groups in the experiment (may account for difference between groups).

Your research assistants start to like your research participant in a study designed to help folks to stop smoking. They gradually change their criteria to what is considered smoking from simply putting the cigarette in one’s mouth to actually breathing in smoke. This makes it look like the DV had decreases

when it has not. This is called:

450 of 35

0% 0% 0%0%0%0%

1. Statistical regression

2. Mortality problem

3. Maturation problem

4. History problem

5. Selection problem

6. Instrumentation problem

You are measuring the effects of caffeine on people who experience panic attacks. One of your caffeine

groups is “run” over the fourth of July weekend. This could be an example of….

1. Statistical regression

2. Mortality problem

3. Maturation problem

4. History problem

5. Selection problem

6. Instrumentation problem

Participant Effects

C7-3a. Be able to list and define the participant effects (demand characteristics, positive self-presentation) the experimenter effects (experimenter attributes, experimenter expectancies). In the section labeled experimenter expectancies, know the different effects on the experimenter and the participants (240-255)

Participant effectsA. Demand characteristics: cues which

give participant clues as to the nature of the experiment and they change their behavior.

B. Positive self-presentation: participants try to appear positive (and often use demand chars. in this quest!)

kjfkjsdkfjksdjf

10%

90%1. Appropriate

2. Inappropriate

Experimenter Effects

C7-3b. Be able to list and define the participant effects (demand characteristics, positive self-presentation) the experimenter effects (experimenter attributes, experimenter expectancies). In the section labeled experimenter expectancies, know the different effects on the experimenter and the participants (240-255)

• Experimenter effects:Attributes: the characteristics or skills of

experimenter may influence results. (gender, method of delivery, etc.)

Experimenter Expectancies: influence of the experimenter’s expectancies on outcomes.– Recording Bias– In interpreting data– Affect participants’ responses

More Extraneous vars

C7-4. Know what a sequencing effect is, as well as subject sophistication (255-257).

More extraneous variables:

• Sequencing effect: prior treatment affects subsequent conditions

• Participant sophistication: being familiar with subject matter or with materials/procedure (more of a problem with repeated measures designs)

Randomization

C8-1. Know what Randomization is, what it accomplishes (mention 'unknown sources of variation')(264), and how to do it (exhibit 8.1, page 266). What does "representative" mean? (267).

Randomization: The only known technique for controlling unknown sources of variation; every member of a population has an equal chance of being selected. (participant variables/extraneous variables)

Representative (Random selection): The extent to which a sample is similar in composition to a population.

Random Assignment should: Distribute potential extraneous variables equally to various groups.--by chance may get bias--the larger the sample, the less likely bias will occur

Random Selection and Assignment

C8-2. What does Random Selection (sampling) assure, and what does Random Assignment Assure? (267,0)

Random Sampling (selection) assures a representative sample

Random Assignment assures a equal distribution of extraneous variables (participant variables) thus making your control and experiment groups equal in those potential confounding variables.

More Extraneous vars

C8-3. Be able to define matching and know the two benefits (273-274,0)

Matching: equating groups on one or more variables by measuring participants on those variables and assigning them in equal amounts to the various groups.

ADV.increases sensitivity of experiment (the DV)More balanced groups

DisADV: time-consuming; pre-measuring induces demand characteristics

Precision Matching: each participant is matched to another on some variable.

ExampleStep 1: Measure on variable suspected of having impactStep 2: put in orderStep 3: Block according to number of groupsStep 4: RA participants from a block to one group – repeat this for each block.

Matching

C8-4. Be able to define matching and know the two benefits (273-274,0)

Three groups – IV = Rock music, Classical music, no musicDV = performance on a math testMatching Variable: IQ scores

11210113211710412413611899113140106

Controlling Subject Effects

C8-5. Know and be able to briefly describe the five general methods for controlling subjects effects (Double-blind placebo, deception, disguised experiment, independent measure of the dependent variable, and procedural/control of subject interpretation)(291-297)

Four General Methods for Controlling Subject Effects1) Double-Blind Placebo: Experimenter and participant unaware of

the conditions in effect (but aware they are in an experiment).2) Deception: Misdirect participants as to the nature of the

experiment. 3) Disguised Experiment: Participant is not aware they are

participating in an experiment.4) Independent Measures of the DV: Measure the DV away from the

experimental condition.5) Procedural control/control of participant interpretation – gain

insight into participants perceptions:A. Retrospective verbal report (exit interview)B. Concurrent Verbal Reports (think aloud)C. Sacrifice groups (terminate at various times)

Control Recording Errors

C8-6. Know the various methods for control of recording errors:

Maintain observer awareness—continuous training

Multiple observers—inter-observer agreements

“Blind” data recorders—unaware of conditions in effect

Automation—computer/videos, etc.

Control Recording Errors

C8-7. Be able to list, describe, or to apply any of the three "Control of Experimenter Expectancy Errors

1. Blind Technique – experimenter unaware of conditions (reduces bias)

2. Partial blind technique – keep experimenter “blind” as long as possible (reduces bias)

3. Automation – use computer, video tapes, audio recordings – minimizes errors

A micro-switch in the base of little Johnny's chair allows us to detect if he gets off his seat. The experiment is investigating if caffeine

increases these out of seat episodes. The teacher is unaware if Johnny had caffeine or placebo. This is an example of…

20%

40%

0%

40%

0%

1. Automation

2. Blind technique

3. Multiple observers

4. Number 1 & 2

5. Nada!

Faulty Designs

C9-1. Given a scenario which describes any of the faulty research designs, you should be able to point out the faults (e.g. potential confounds). There are three type of faulty experiments described (313-317)

A. 1-group after-only design: one group is measured on the DV after IV is manipulated/administered did a change in performance occur? For what reasons?

B. 1-group before-after design (pretest/posttest): change could be due to history/maturation/instrumentation/stat regression

C. Non-equivalent posttest only: could be differences between groups causing effects (ex. Joiners vs. nonjoiners) too small of a sample or not randomly assigned

True Research Design

C9-2. Be able to list and describe the three

criterion that a true research design must meet:

A. Adequate answering of question/test of hypothesis (317,1)

B. Control of extraneous variables (318,1)C. Allows for generalizability/external validity

(319,3)


C9-3. Know what a control and experimental group is, and the two functions that a control group serves (318,4-319,1)

Experimental group—receives some level of the IV

Control group receives no IV or some fixed/traditional levelControl group allows:

1. Comparison of scores with experimental group

2. “Control for rival hypotheses” meaning detect extraneous variable influences. (detect by looking at the changes…if DV changes also, then confound)


C9-4. Be able to list and briefly describe the five reasons a pretest might be incorporated into a study. (320-322). Also know the two disadvantages of the pretest (322,3)

Pretest might be incorporated for:1) increased sensitivity/matching2) to determine if ceiling/ floor effect 3) get initial position/attitudes4) similar initial comparability: compare subjects5) evidence of change (after posttest) Disadvantages

ExpensiveDemand characteristics

After Only Designs

C9-5. There are six "After-Only Designs" described in your text (323-341). Be able to describe each and list the advantages and disadvantages. Do the same for the Before-After Design described in 339-341.

1 Between groups/participants after-only design (diff. parts. in each group)

a. Control and experimental groupb. Random selection and random assignment

Disadvantage:Randomization could be biased (likelihood is a function of sample size)Not matched = less sensitive design

After Only Designs

Simple Randomized Participants DesignA. Control and experimental groupB. Random selection and random

assignmentC. More than one level of the IV

After Only Designs

Factorial Designa. Control and experimental groupb. Random selection and random assignmentc. More than one level of the IV and more than one

IV Can determine:

Main effect: the effect of each IV taken aloneInteraction effect: the effects one may have on DIFFERENT levels of another IV

Know: 2 x 22 x 3 x 22 x 4 x 6 x 2 (levels, # of variables, total number of

participants)

A 2x2x5 factorial design has

:400 of 35

0% 0%0%0%

1. 1 IV

2. 20 Ivs

3. 3 Ivs

4. 9 IVs

A 3 x 6 factorial design has

:400 of 35

0% 0%0%0%

1. 18 groups

2. 9 groups

3. 2 groups

4. 36 groups

A 2 X 2 design could be

:400 of 35

0% 0%0%0%

1. 0, 1, and 2 mg of caffeine and gender

2. Gender and room color: red, blue, green orange and white

3. Gender and a hot room or a cold room

4. Caffeine or not, and room color: blue, red, green.

A 2 x 2 design, with 20 people per group, has

:400 of 35

0% 0% 0%0%0%

1. 24 participants

2. 80 participants

3. 40 participants

4. 100 participants

5. 20 participants

After Only Designs

C9-6. In regard to a factorial design (328-335), know what a cell refers to; what the main effect and interaction effects are; and be able to recognize graphical displays of interaction effects (e.g. if they exist or not see figure 9.10)

Interpreting factorial design results1) If only main or only interaction effects, interpret both 2) If main and interaction effects: interpret only interaction

effects Problems with factorial:1) More factors = lots more participants2) Hard to manipulate more than 2 IVs3) Triple interactions and more = hard to interpretAdvantages of Factorial1) Test more (more IVs)2) Can build extraneous variables into experiment (like

gender…crowding…aggression levels)Can examine interactions (also more like real world)

After Only Designs

4) Within-groups/participants after-only design (repeated measures)

all participants get all treatments

Advantages: don’t have to create equivalence on participant

variables fewer participants

Disadvantages: sequencing effects (who goes first/second) control with counterbalancing no control (detection) of history/maturation, etc.

After Only Designs

Combining Between and Within Subject Designs(a factorial design)

More Variables studiedLess subjects than a pure between designCan test for interactions IV 1

IV 2

S1 S4

S2 S5

S3 S6

S1 S4

S2 S5

S3 S6

S7 S10

S8 S11

S9 S12

S7 S10

S8 S11

S9 S12

Single Subject Designs

C11-1. Know what a single-SUBJECT design is in general (be able to describe its components) (373-377). Add to this that seldom is a single participant only used—usually 5-6 is used, providing replication evidence for any observed effects (lecture).

Examine one participant at a time to investigate the

effects of an IV (treatment).a. Repeated measuresb. “Baseline as a control and comparison”c. Use multiple participants = replications


C11-2. Be able to list and describe each of the single-participant designs listed (four are described). Know how each attempt to demonstrate that the independent variable was responsible for any effects observed (377-395).

ABA designBaseline (A) – Treatment (B) – Baseline (A)Withdrawal lends evidence that treatment caused effect!

ABAB designMore evidence!

Interaction Designs: Interaction refers to having two IVs

present at the same time thus:

A B A B A BC A B A……..


DV

Baseline - A Treatment - B Baseline - A

Be able to create an ABA, or ABAB graph



Multiple Baseline designs (across participants or situations/locations)

Baseline is recorded during the same time period for different lengths of time, for multiple participants. The treatment (IV) is introduced at different times for each participant (baseline is staggered/switch to IV is at different times for each participant to gauge the effect of IV)

No change in ongoing baseline is evidence that extraneous variables are NOT at work



Multiple Baseline designs

From: http://curry.edschool.virginia.edu/class/edlf/733/acts/mbline.html


Changing criterion designs

• If effects follow the change in the criterion (the IV) then evidence supports the IV causes effect in the DV.


Changing criterion designs

From:http://www.radford.edu/~pjackson/2SmallN.pdf


• Know the methodological considerations in using single-participant designs. Under the heading “Baseline,” I found four points regarding a stable baseline (absence of a trend; trend opposite to treatment effect; excessive variability; and reactivity). Know the critical issues related to changing one variable at a time (only change one!) (395,3-398).

• Stable baseline is characterized by:– Absence of trend and only slight degree of variability

(e.g. w/in 5% mean)– Trend opposite, treatment is powerful enough to

produce effect and reverse previous– Extreme fluctuations: check all components of the

study and try to identify and control sources of variability

– Reactivity: fact of using human participants—taking data itself may effect behavior


Knowing when to change from baseline to intervention:

1) Shouldn’t change baseline until observe:

Little variability

No trends

(but could if trend is opposite in direction to expected effect)

2) Only change one variable at time!

(Examples on board)


Knowing when to change from baseline to intervention:

1) Shouldn’t change baseline until observe:

Little variability

No trends

(but could if trend is opposite in direction to expected effect)

2) Only change one variable at time!

(Examples on board)

Scales of Measurement

S1-1. Scales of Measurement: important for selecting stat's (later on)

1. Nominal Scale: number is really a name!1 = male 2 = female

2. Ordinal Scale: number represents degree of an attribute: (order things!)take all exams and rank: 1,2,3,4,5,but 5-1 = 4 no information!

3. Interval Scale:A. different degrees of an attribute are indicatedB. different levels, or degree numbers are equally spacedC. zero point is arbitrary (doesn't indicate absence of variable)

4. Ratio Scale:A. numbers are equidistant (as in interval)B. numbers indicate diff. degree of attributeC. O indicates absence of attribute

Which one is nominal?

450 of 5

1 2 3 4 5

20% 20% 20%20%20%1. The number of cigarettes a person

smokes in a day

2. Level in school: freshman= 1, sophomore=2, etc.

3. Age in years

4. How happy are you?

none 0 ------5 --------10 Lots

Which one is ratio?

450 of 5

1 2 3 4 5

12% 12% 12% 12%12%12%12%12%1. Blood pressure level

2. Heart beat rate

3. Age in years

4. Number of dates one had last week

5. The number of hours of TV watched in a month

6. All of the above

7. None of the above

8. I haven’t the faintest!

PSYC 3000. C1-1 Know the five steps involved in the scientific method. “Know” means to be able to list and briefly describe each step (14-18). –ID problem:

Documents

discoverability slide

intelligent slide

personal biasopinion

scientific approach

objectives of science

descriptive research

characteristics of science

laws of nature