CAS 301 Exam #2 (Chapters 6-10)

Counterbalancing

Allows researcher to determine the extent to which order influences results.

Variable

Any event, situation, behavior, or individual characteristic that varies. Must have 2 or more levels.

How strong should the manipulation of your IV be?

As strong as possible while keeping the experiment realistic.

Types of questions

Attitudes and beliefs, facts, and demographics, behaviors

External Validity

Generalizability

Systematic Observation

Observation of one or more behaviors in a particular setting. -Interested in only a few behaviors, NOT the whole setting -Typically coding systems are developed to help categorize behaviors -Hypotheses are usually formed before completing the observation. -Ex: watch children play with each other-does a child: watch others, play alone, play together, parallel play, group play -Watch out for reactivity, reliability of measures, and sampling

Latin Squares

All conditions both precede and follow all other conditions.

Randomization

All participants are equally likely to be in either experimental condition-ensures that the extraneous variables are as likely to affect one group as the other.

Construct Validity

(6 types) Face, content, predictive, concurrent, convergent, discriminant, predictive

Non-Experiments

(Correlational design) Relationships are studied by making observations or measures of the variables of interest-BOTH variables are measured.

Main Effects

-A main effect is the effect that each variable has by itself. -To find main effect, add the scores in each level of the IV, across both levels of the other. -For example, to determine the main effect of type of education, add the scores for each level of education across the other IV (development). If answers are different from another, then there is a main effect. -The bigger the difference between the 2 (or more) levels within one IV, the more likely there is a significant main effect of that variable.

Independent Groups Design

-Also known as between-subjects. Compares scores across participants. -Ex: You design an experiment with two groups. You compare the scores of group 1 to group 2 -Participants are assigned to conditions using random assignment, which allows the groups to be considered equivalent. -Should be used when your experiment produces a permanent, change in participants removal of brain tissue.

Factorial Designs

-Always have at least 2 IVs with at least 2 levels in each. -Number of IVs and levels are represented through numbers.

Interactions

-An interaction means that the effect of one IV is different at different levels of the other IV. -The effect of one independent variable depends on the level of the other independent variable. -Interactions are also called moderator variables. -Moderator variable influence the relationship between the other IV and the DV.

Demand Characteristics

-Any feature of an experiment that might inform participants of the purpose of the study. -Can reduce this with deception (creating cover stories) or by hiding the measure in a set of filler items to throw participants.

Pros of using a pretest

-Can asses if groups were equivalent before manipulating IV -Necessary whenever there is a possibility that participants will drop out of experiment (attrition/morality) because groups might become nonequivalent during the experiment and you need a pretest to show that to you.

Dillman's Principles

-Choose simple over specialized words. -Choose as few words as possible to pose the question. -Use complete sentences to ask questions. -Avoid vague qualifiers when more precise estimates can be obtained. -Avoid specificity that exceeds the respondent's potential for having an accurate, ready-made answer. -Use equal numbers of positive and negative categories for scalar questions. -Distinguish undecided from neutral by placement at the end of the scale. -Avoid bias from unequal comparisons. -State both sides of attitudinal scale in the question stems. -Eliminate check all that apply question formats. -Develop response categories that are mutually exclusive. -Provide accurate time referents. -Avoid double-barreled questions. -Soften the impact of potentially objectionable questions. -Avoid asking respondents to make unnecessary calculations.

Measuring the DV

-DV should be sensitive enough to detect differences between groups. -If a measure is not sensitive enough to detect differences, then there is no way to know if the IV affects the DV. -The ceiling and floor effect make it seem like the IV doesn't influence the DV because most participants score the same way. -The measure is not sensitive enough to detect differences as a result of the IV.

Behavioral Measures

-Direct observations of behaviors. -Important to note the rate, reaction time, and duration of behaviors.

Advantage of the Solomon Four Group

-It can test for the impact of the pretest-if there is no impact of the pretest, then the posttest scores will be the same in the two control groups and in the two experimental groups -It can also determine that groups were equivalent at start of experiment (at least for the groups that received the pretest)

Repeated Measures Design

-Known as within-subjects -Compares same people at more than one point in time. -Participants are repeatedly measured on the DV after being in each condition of the experiment. -Need fewer participants than Independent Groups Design -Easier to discover systematic individual differences from the effect of the IV -Watch out for: practice effects, fatigue effect, carryover effect, and order effects

3 Main types of results from factorial designs.

-Main effects -Interaction -No relationship

Staged Manipulations

-Manipulate the IV by staging events during the experiment. -Often uses confederates. -Most likely to be used in one of two events: when the researcher wants to create some psychological state in participants (anger, elation, etc.) or when simulating a real-world occurrence.

Straightforward Manipulations

-Manipulate variables with instructions and stimulus presentations. -Info may be presented verbally, in written form, through video, or computer.

Matched Pairs Design

-Match participants on a 'participant variable' (age, sex, etc.) or a variable that is strongly related to the DV -Used when access to a large sample size is limited -Have a greater ability to detect a significant effect of the IV because you can account for individual differences in response to the IV

Naturalistic Observation

-Observations of individuals based on their natural environments. -Goal is to provide complete, accurate picture of a setting, but can also include specific behaviors (aggression in bars). -Called 'field work' field observation -very time consuming- which means that this can be difficult to do -can either involve participant or non participant observation

Physiological Measures

-Recordings of responses of the body. -Ex: galvanic skin response, EMG, EEG, MRI, fMRI.

Self-report Measures

-Used to measure attitudes, beliefs, behavior, past behavior, etc. -Scales are most commonly used.

Expectancy Effects

-When experimenters develop expectations about what results should be obtained from an experiment. -Solutions: Train experimenters to be unbiased, run all conditions simultaneously, automate procedures, single blind and double blind experiments.

Placebo Effect

-When the results of the experiment may not be due to the IV, but instead result from participants expectations -Balanced placebo design

How to communicate research?

-conferences/professional meetings -publishing in journals -blogs, social media, websites, facebook, etc.

Observed Score

= True Score + error (random and systematic)

Correlation Coefficient

A numerical index of the strength of relationship between two variables. Ex: Pearson's Product-Moment Correlation

Internal Validity

Cause and effect (Need temporal precedence and covariation of cause and effect, elimination of alternative explanations)

Third Variable Problem

Confounding variables

Case Studies

Describe an individual person (or setting: school, neighborhood, etc.) -Good for observing things that are rare or unusual

Non probability sampling

Do not know the probability of any particular individual of the population being selected to participate

Probability Sampling

Each member of the population has a specific probability of being chosen

Interrater

Extent to which raters agree in their observations

Quantitative Data

Focuses more on breadth-want more people and specific, quantifiable behaviors. Conclusions drawn from statistics.

Experiments

Involve direct manipulation and control of the IV. Can establish cause and effect. With experimental control, all extraneous variables are kept constant, If a variable is held constant, it cannot be responsible for the effects of the experiment.

Test-Retest

Measuring the same individuals at two or more points in time

Disadvantages of the Solomon Four Group

Need a lot of participants because instead of having 1 experimental and 1 control group, there are now 2 experimental and 2 control groups (4 groups total, hence the name of the design) -Rule is that you need at least 30 people in each group, but most likely you will need more.

Scales

Nominal, ordinal, interval, and ratio

Main difference between non probability and probability sampling

Non probability sampling does not involve random selection and probability sampling does

Types of Relationships

Positive linear, negative linear, curvilinear, no relationship

Cons of using a pretest

Pretest Sensitization= participants can go on autopilot and can guess hypothesis

Psychobiography

Researcher applies psychological theory to explain the life of an individual.

Non-participant Observation

Researcher is an outsider to the setting or group studies. -This created issues of acceptance and reluctance to share info from group members. If the researcher is an outsider to their group, they will be less willing to openly share all information.

Participant Observation

Researcher is fully immersed in the setting. The researcher joins the group that they observe. -Problem: potential for bias

Randomization controls for SELECTION DIFFERENCES

Researchers must ensure that people selected to be in the conditions cannot differ systematically in any way

Square

Row 1: A B D C Row 2: B C A D Row 3: C D B A Row 4: D A C B

Response Set

Tendency to respond to all questions from a particular perspective that may or may not accurately reflect answers to the questions themselves. Ex: Social Desirability and Yea-Saying or Nay-Saying

Internal Consistency

The consistency between items on the same scale

Operational Definition

The set of procedures used to measure or manipulate the variable. It is needed to study variable empirically. It also helps researchers communicate their ideas with each other and with others because they can discuss abstract ideas in concrete terms.

Qualitative Data

Typically an observation of the world around us that focuses on people behaving in natural settings. For instance, if you want to observe social interaction at a school-sponsored event, you could sit somewhere during the even and record all aspects of the event and social interactions. This data collection involves a lot of writing and text analysis. Focus on depth of data-want data that is rich- describes the phenomenon in great detail. Conclusions can be drawn by investigator.

Archival Research

Uses previous research to answer research questions (ex: Statistical records, survey archives, mass communication records)

Floor Effect

When a task is so difficult that hardly anyone can perform well.

Graphing Factorial Designs

When graphing factorial designs, if there is a main effect of the variable, the lines are parallel. If there is an interaction, the lines are not parallel.

Ceiling Effect

When participants quickly reach the maximum performance level.

Survey question

ex: Halloween makes me feel excited and nervous What principle is violated with this question?

Types of non probability

haphazard, purposive, quota

Types of responses participants can give

open ended or close ended

Experimental Design

post-test only, pre-test post-test, Solomon four group

Types of probability

simple random, stratified random, cluster