Please enable JavaScript.

Coggle requires JavaScript to display documents.

Empirical Methods & Experiment Design - Coggle Diagram

- - - - Steps
        
        Form a hypothesis
        tentative answer or explanation
        
        Use hypothesis for a testable prediction
        
        Evaluate prediction by
        systematic planned observations
        
        Use observations to support, refute,
        refine original hypothesis
        
        Observe behavior or phenomena
      - Process of Scientific Inquiry
        
        observation
        
        hypothesis
        
        prediction
        
        planned observation
        
        new hypothesis
        
        prediction
        ...
    - - Inductive
        increase
        small set of observations as basis for a general statement about a larger set of possible observations
      - Deductive
        decrease
        a general statement as the basis for a conclusion about specific examples
    - - Hypothesis
        a statement that describes or explains
        a relationship between variables
        → can lead to several observable/measurable predictions
      - Variables
        characteristics or conditions that change or have
        different values for different individuals
      - Theory
        set of statements about the mechanisms underlying a particular behavior. Theories help organize and unify different observations of the behavior and its relationship with other variables.
        A good theory generates predictions about the behavior
    - - Qualitative Research
        
        narrative reports
      - Quantitative research
        
        statistical analysis
      - Steps in research Process
        
        1. Find research idea
        
        select general topic
        
        review literature in area
        
        identify relevant variables
        
        find unanswered questions
        
        Mantra
        
        Interest
        
        broad background information
        
        open mind towards more specific topics
        
        focus on specific questions
        one step at a time
        
        Sources for Ideas
        
        Personal interests & curiosities
        
        Casual Observation
        
        Reports of others observations
        
        informal sources
        
        trustworthy sources
        
        Practical Problems / questions
        
        Applied Research intends to answer practical questions
        
        Basic research/fundamental research intends to answer theoretical questions or gather knowledge simply for its own sake
        
        Behavioral Theories
        
        checking whether a theory is true in a certain setting
        
        checking for opposing theories
        
        Background Literatures
        
        Primary Sources
        firsthand reports of observations / research results written by the individual who actually conducted the research and made the observation
        
        Secondary Sources
        description / summary of another persons work.
        written by someone who did not participate in research/observations being discussed
        
        Preparing Literature Search
        
        find keywords:
        
        textbook headings/subheadings
        
        note subject keywords
        
        Author names
        
        Online Database
        
        summaries
        
        abstracts
        
        is a brief summary of the publication usually about 200 words
        
        full-text
        
        Weed Out irrelevant material
        
        title
        
        abstract
        
        references from found articles
        
        Define Research Idea
        
        explicit statements in articles regarding
        
        unanswered questions
        
        limitations of studies
        
        conflicting results
        
        Critical Reading
        
        Components of an Article
        
        Introduction
        
        literature review
        
        hypothesis or purpose for study
        
        specific prediction from hypothesis
        
        Method
        
        participants
        
        Procedure
        
        Results
        
        Statistics
        
        Discussion
        
        results on hypothesis
        
        justified conclusions
        
        alternative explanations
        
        applications
        
        limits to generalization
        
        References
        
        list of items cited
        
        Issues in every step of an article
        
        2. Form hypothesis
        
        Characteristics
        
        Logical
        
        logical conclusion of logical argument
        
        rational method
        
        Testable
        
        a testable hypothesis is one for which all variables, events and individuals can be defined and observed
        
        Refutable
        
        a refutable hypothesis is one that can be demonstrated to be false. That is, it is possible fot the outcome to be different from the prediction
        
        Positive
        
        must make a positive statement about the existence of something
        
        a hypothesis that denies existence cannot be tested in a research study -> no good foundation of research
        
        3. Determine how to define
        & measure variables
        Transfer of hypothesis into an empirical testable form
        
        Variables
        characteristics or conditions that change or have
        different values for different individuals
        
        abstract entities
        
        different options for measurement
        
        Types of variables
        
        well-defined
        
        easily observed
        
        easily measured
        
        abstract
        
        intangible
        → complicated measurements
        
        Realtionships
        
        one-to-one
        
        Constructs and Operational Definitions
        
        Constructs
        hypothetical attributes or mechanisms that help explain and predict behavior in a theory
        External Stimulus -> Construct -> External Behavior
        Reward -> Motivation -> Performance
        
        Operational Definitions
        methods of defining and measuring constructs
        
        intelligence can be operationally defined as a score on an IQ test
        
        "hunger" can be operationally defined as the number of hours of food deprivation
        
        Limitations
        
        operational definitions is not the same as the construct itself
        
        concerns on quality of operational definitions and measurements produced
        
        could leave out of important components of a construct
        
        include extra components that are not part of the construct being measured
        
        Criteria for Measurements
        
        Reliability & Validity
        No Reliability -> No Validity
        Validity -> Reliability
        Consistency of measurement is no guarantee of validity
        (broken stopwatch measures reliable but not valid)
        Reliability -/-> Validity
        It is not necessary for a measurement to be valid for it to be reliable
        No Validity -/-> Reliability
        
        Reliability
        is the measurement of the procedure stable. If the same individuals are measured under the same conditions, a reliable measurement produces (nearly) identical measurements
        measured score = True score + Error
        
        Errors
        
        3 more items...
        
        Types of Reliability
        
        3 more items...
        
        Validity
        Does the procedure measure what it claims to measure?
        the validity of a measurement procedure is the degree to which the measurement process measures the variable that it claims to measure
        Establish Validity:
        
        measurement procedure must accurately capture the variable that it is supposed to measure
        
        Common Definitions of Validity
        
        4 more items...
        
        Measured by
        Consistency of relationship of two sets of measurements
        Correlations
        
        positive relationship
        
        measurements change together in same direction
        
        negative relationship
        
        measurements change in opposite directions
        
        Consistent positive relationship = correlation near +1
        
        consistent negative relationship = correlation near -1
        
        Inconsistent relationship = correlation near zero
        
        Process of Measurement
        components:
        
        set of categories
        
        procedure to assign individuals to categories
        
        Scales of Measurement
        distinction between ordinal & interval is not always clear
        
        Nominal Scale
        REVEALS WHETHER DIFFERENCES EXIST
        differences in a series of ranks
        represent qualitative (not quantitative) differences in the variable measured
        
        no direction of difference is usable
        
        Ordinal Scale
        INDICATE DIRECTION OF DIFFERNECE
        series of equal intervals with true zero value
        different names for categories, organized sequentially
        
        series of ranks (first, second, third / S,M,L,XL, upper, middle, lower class)
        
        directional relationship between categories
        → not compatible with arithmetic / appropriate statistical analysis
        
        Interval and ratio Scale
        DETERMINES DIRECTION & MAGNITUDE OF DIFFERENCE
        qualitative differences (hound, cat, mouse)
        Organized sequentially with all categories with the same size. Ratio scales have a determined point for zero which is a meaningful point representing none
        → each "step" is the same size
        ratio:
        
        seconds
        
        pounds
        
        interval
        
        Fahrenheit
        
        interval-scores can also be set in relationship to others making them ratio-scores
        
        basic arithmetic
        
        suitable for statistical analysis
        => allow hypothesis testing
        
        Modalities of Measurements
        Issues:
        
        Desynchrony: lack of agreement between two measures
        
        Limit problems:
        
        combine measures into a single scores / individually
        
        Self-Report Measures
        ask direct questions to study behavior
        +
        direct measures
        -
        validity is dubious
        
        biased
        
        honesty
        
        Physiological Measures
        based on manifestations of underlying construct
        +
        objective measures
        -
        
        expensive equipment
        
        may not provide valid measurement of C (different reasons for outcome might be possible)
        
        Behavioral Measures
        based on apparent behaviors that can be observed and measured
        +
        provide researchers with cast number of options (select behavior that fits best)
        -
        behaviors may be only temporary or situational indicators of an underlying construct
        
        Problems
        
        Range Effects
        a measurement that is not sensitive enough to detect a difference
        
        Ceiling Effect
        clustering of scores at the high end of the scale
        
        Floor Effect
        clustering of scores at the low end of the scale
        
        Artifact
        non-natural feature accidentally introduced into something being observed
        
        Experimenter Bias
        → invalid & unreliable results
        Pygmalion-Effekt
        
        Limiting Experimenter Bias
        
        standardize / automate experiment (limit personal contact between experimenter & participant)
        
        single-blind: if the researcher does not know the predicted outcome / double-blind: if both the researcher and the participants are unaware of the predicted outcome
        
        Demand Characteristics
        cues that
        
        suggest to participants what the purpose and hypothesis is
        
        influence participants to respond / behave in a certain way
        
        Participant Reactivity
        occurs when participants modify their natural behavior in response to the fact that they are participating in a research study or the knowledge that they are being measured
        
        Field Studies /covert field study → +
        Laboratory studies → -
        
        Role Adoption
        
        good subject role
        
        negativistic subject role
        
        apprehensive subject role
        
        faithful subject role
        
        4. + 5. Participants
        
        identify participants for study
        
        decide how to select
        
        plan ethical treatment
        
        Ethics in Research
        research ethics concerns the responsibility of researchers to be honest and respectful to all individuals who are affected by their research studies or the reports of the studies' results
        // no issue of morality
        
        What to be concerned about
        
        what measurement techniques may be used
        
        hot to select participants
        
        which research strategies to use
        
        which research design to use
        
        How to carry out the study
        
        How to analyse the data
        How to report the data
        
        Basic Principles
        
        Welfare & Dignity of Individuals
        
        Accuracy & honesty of research
        
        by
        
        accuracy
        
        honesty
        
        validity
        
        Major Ethical Issues
        
        No Harm
        no comparison should be done when one treatment is considered inferior to the other treatment
        
        Informed Consent
        
        Consent form
        
        requires the investigator to provide all available information about a study so that an individual can make a
        rational, informed decision to participate in the study (before beginning the study).
        
        Debriefing
        
        explanation what the study was really about
        
        counteracting or minimizing negative effects
        
        explaining nature of and justification for deception
        
        answering questions the participant has
        
        Deception
        purposefully withholding information / misleading participants with regard to information about a study
        
        active
        
        misinformation of the prticipants
        
        passive
        
        withholding information intentionally
        
        Confidentiality
        Ensured through anonymity
        
        attitudes & opinions
        
        measures of performance
        
        demographic characteristics
        
        Fraud
        explicit effort of a researcher to fasify or misinterpret data
        
        prevent fraud
        
        Replication
        replication of research using the same procedure used in the original. Either the replication supports the original study by duplicating the original results or it casts doubt on the original study demonstrating that the original study is not easily repeated
        
        Plagiarism
        unethical representation o someone else's ideas or work as one's work
        
        Sampling
        goal: generalize from sample to larger population
        
        Population
        entire set of individuals of interest to a researcher. Although the entire population usually does not participate in a research study, the results from the study are generalized to the entire population
        school-children, adolescents interest in gaming
        == all individuals of interest
        
        Accessible Population
        portion of target population consisting of individuals who are accessible to be recruited
        
        Sample
        set of individuals selected from a population and usually is interpreted to represent the population in a research study
        share the same defining characteristics with the population
        
        Representativeness
        extent to which the characteristics of the sample accurately reflect characteristics of population
        the more representative the better to generalize
        
        accessible population = target population?
        
        Threats
        
        Bias
        biased sample = sample with different characteristics from those of the population
        Selection bias / sampling bias = occurs when participants are selected in a manner that increases the probability of obtaining a biased sample
        
        Sample Size
        Law of large number = the larger, the higher the probability of representation
        -> 25-30 individuals for each group/treatment condition
        
        Sampling Techniques
        
        Probability Sampling
        
        entire population is known
        
        each individual on population has a specifiable probability of selection
        
        sampling occurs by a random process
        
        requires knowledge on whole population
        
        Random Process
        the outcome must be unpredictable each time / process must guarantee that each o the possible outcomes is equally likely to occur
        
        Probability Sampling Methods
        
        6 more items...
        
        Nonprobability sampling
        population is not completely known, individual probabilities cannot be known and the sampling method is based on factors such as commonsense or ease, with an effort to maintain representativeness and avoid bias
        
        odds of selecting a particular individual are unknown
        
        → greater risk of biased samples
        
        Convenience Sampling
        individual participants are obtained by selecting those who are available and willing
        +
        easy method
        
        interested participants are reliable
        
        less expensive
        
        -
        weak form of sampling
        
        no attempt to know the population
        
        no attempt to randomize process
        
        → biased
        
        Quota Sampling
        subgroups are identified to be included
        quotas are established for individuals to be selected through convenience from each subgroup
        +
        
        control over composition of convenience sample
        
        -
        
        biased sample (same subgroup → same socio-economic background → biased)
        
        convenience sample
        
        5. select research strategy
        
        Experimental Strategy
        an experiment (true experiment) attempts to show that changes in one variable are directly responsible for changes in a second variable
        
        Cause & Effect
        
        Problems
        
        Directionality Problem
        relation between two variables does not always explain direction
        
        Confounding Variables
        
        3 more items...
        
        Third Variable Problem
        relation between variables does not necessarily mean that there is a direct (causal) relationship
        
        Goals
        
        Cause before effect
        
        responsibility of relation
        
        Basic Elements
        
        Manipulation
        a manipulation consists of identifying the specific values of the independent variable to be examined and then creating a set of treatment conditions corresponding to the set of identified variables
        
        ...of independent variable
        
        1 more item...
        
        Manipulation for Cause & Effect
        
        Measurement
        a second dependent variable is measured for a group of participants
        
        Comparison
        the scores of both treatment conditions are compared
        
        Control
        
        Confounding Variables
        
        control extraneous variables
        
        eliminate confounding variables
        
        3 more items...
        
        Control Group
        no-treatment condition in an experiment
        (control condition if same group in different condition)
        
        2 more items...
        
        Manipulation Check
        additional measure to assess how the participants perceived and interpreted the manipulation and/or assess the direct effect of the manipulation
        
        4 more items...
        
        Research Strategy
        is a general approach to research determined by the kind of question that the research study hopes to answer
        
        Category 1
        Descriptive Research
        
        Descriptive Research
        Category 1
        
        not concerned with relationships between variables
        
        obtains a snapshot
        
        data in averages/percentages
        
        numerical: calculate the mean
        
        non-numerical: percentages for each category
        
        Category 2
        Correlational Research
        
        Correlational Research
        Category 2
        
        describes relationship, not causation
        
        measures multiple variables per individual
        
        one group of participants -> measures two variables for each individual
        analysis:
        
        numerical: correlation calculation
        
        non-numerical: chi-square test
        
        Scatter Plot
        
        each individual is representad by a point
        
        Relationships between Variables
        
        Linear
        
        Curvilinear
        
        Positive
        
        Negative
        
        Category 3
        Comparing Two or More Sets of Scores
        analysis:
        comparing groups of scores
        
        t test
        
        chi-square
        
        variance analysis
        
        Quasi-Experimental Research
        produces a cause-and-effect explanation using preexisting participant characteristics or time
        used for cases where you cannot assign groups
        
        can never produce an unambiguous explanation
        
        Ex.: does treatment vs. no treatment cause a reduction in smoking?
        
        -
        no random distribution of participants → no experimental study
        
        Non-Experimental Research
        
        demonstrates a relationship between variables
        
        causation (explanation)
        
        compares two groups of scores (measures only one variable for each individual) / two groups OR two different times
        -
        no random assignment → no experimental study
        
        no control group / multiple measures → no experimental study/ quasi experiment
        
        Experimental Research
        
        cause-and-effect relationship
        
        Ex.: Are increases in exercise responsible for causing decreases in cholesterol level?
        
        Validity
        impossible for a single research to eliminate all threats to validity
        → balance internal & external validity
        
        Internal Validity
        if the research produces a single unambiguous explanation for the relationship between two variables
        
        1 more item...
        
        External Validity
        extent to which results can be generalized to people, settings, time, measures and characteristics other than used in that study
        
        2 more items...
        
        threats to both
        Internal and
        external validity
        
        2 more items...
        
        Research Design
        Decisions on
        
        group versus individual
        
        within group or between groups
        
        same individuals versus different individuals
        
        number of variables to be included
        
        Research Procedure
        research procedure is an exact step-by-step description of a specific research study
        
        exact step-by-step description of a specific research study
        
        Determination of
        
        how the variables are to be manipulated, regulated & measured
        
        how many individuals will be involved
        
        how the individual participants will proceed through the course of the study
        
        6. select a research design
        SESSION 8+9
        
        decide on specific methods and procedures
        
        ⇒ basis for different designs of studies
        
        Between Subjects Design
        independent measures experimental design requires a separate, independent group of individuals
        -> one score per individual
        -> separate groups for different treatment conditions
        
        Drawbacks
        
        Threats to Internal Validity
        
        Assignment Bias
        exclusively for research with a between-subject design
        
        groups with different characteristics threaten the validity of study
        
        2 more items...
        
        Confounding Variables
        
        Individual Differences (→ Assignment Bias)
        
        Environment variables
        
        2 more items...
        
        Differential attrition
        participant dropout rate differs from one group to another
        
        Communication between groups
        
        diffusion (treatment spreads to control group)
        
        compensatory equalization (control group demands the same treatment)
        
        compensatory rivalry (control group changes normal behavior)
        
        resentful demoralization (control group becomes less motivated)
        
        Large Number of Participants
        
        Benefits
        
        No Training Effect
        
        No Fatigue / Boredom
        
        Statistical Analyses
        
        Two-Group mean difference
        simple way to compare extremes of the spectrum
        +
        
        simplicity
        
        best opportunity to maximize the difference between the two treatment conditions
        
        opposite extreme values can be selected
        
        -
        
        provides little information
        
        researcher obtains only two real data points for comparison
        
        Limits a study to two groups
        
        Single factor multiple-group design
        More than 2 two groups
        Example: comparing driving performance under three telephone conditions (while talking on the phone, while texting and without a phone)
        → more evidence for true cause & effect relationship
        
        analysis of variance ANOVA
        
        too many groups let the effects become smaller
        
        Comparing Proportions
        = if DV is nominal/ordinal
        Chi-Square Test
        
        Within-Subjects Design
        measures design, compares two or more different treatment conditions (or compares a treatment and a control) by observing of measuring the same group of individuals in all of the treatment conditions being compared.
        "repeated measures design"
        
        Internal Validity
        
        Potential Confounding
        
        Environmental variables
        
        Confounding from time-related variables
        
        2 more items...
        
        Benefits
        
        few participants
        
        eliminates problems based on individual differences
        
        each subject appears in every treatment condition
        
        each subject serves as his own control
        
        reduces variance
        
        Disadvantages
        
        series of condition for each participant
        → influences participant
        
        participant attrition
        
        Statistical Analysis
        
        interval / ratio
        
        repeated-measures t
        
        single-factor ANOVA
        
        ordinal/rank ordered
        
        Wilcoxon test
        
        Sign test
        
        Multiple Treatment-Designs
        
        Withing Subject Design VS Between Subject Design
        Factors:
        
        individual differences
        
        time-related factors & order effects
        
        fewer participants
        
        Matched Subject Design
        each individual in one group is matched one-to-one with an individual in every other group
        
        separate group in each treatment condition
        
        Goal: duplicate advantages of within & between-subjects designs without differences of either one
        https://explorable.com/matched-subjects-design
        
        Advantages
        
        data is more likely to reveal functional relationship
        
        more convincing demonstration of cause-and-effect relationship than a two-treatment design
        
        Disadvantages
        
        too many treatments possible
        
        distinction becomes too small to differentiate
        
        more treatments → more time needed → higher participant attrition
        
        counterbalancing is more difficult as number of treatments increases
        
        Nonexperimental and
        Quasi-experimental Strategies
        involve comparison of scores from different groups/conditions. Non-manipulated variable to define the groups or conditions being compared
        
        the degree to which the research strategy limits confounding variables and controls threats to internal validity
        
        Nonexperimental Design
        between-subjects design
        
        little or no attempt to minimize threats to internal validity
        
        Purpose
        compares preexisting groups of individuals
        Examples of design
        
        differential research
        
        posttest.only nonequivalent control group design
        
        pretest-posttest nonequivalent control group design
        
        cross-sectional developmental design
        
        Between Subjects Non-/Quasi-experimental Designs
        
        Nonequivalent group design
        -preexisting groups-
        different groups of participants are formed under circumstances that do not permit the researcher to control the assignment of individuals to group
        Goal: balance/equalize groups
        Mean: random asssignment process
        
        2 more items...
        
        Single-variable research
        
        quasi-experimental
        
        Correlational
        
        Qualitative research
        
        Within-Subjects Non-/Quasi-Experimental Designs
        
        Non-Experimental Pre-Post
        Non-Experimental
        series of observations is made over time for one group of participants
        Threats to Internal Validity
        
        History
        
        Instrumentation
        
        Testing Effects
        
        Maturation
        
        Statistical Regression
        
        Time-Series Design
        Quasi Experimental
        "interrupted time series design"
        series of observations for each participant before a treatment or event and a series of observations after the treatment or event
        
        a manipulation administered by the researcher
        
        an outside occurrence that is not controlled or manipulated by the researcher
        
        Limits Threats to Internal Validtiy
        
        pretest observations
        
        identify trends before treatment of
        
        maturation
        
        instrumentation
        
        regression
        
        1 more item...
        
        Developmental Research Designs
        examine changes in behavior related age
        
        Cross-Sectional Developmental Design
        Between-Subjects Design
        Non-Experimental
        different groups of individuals, a group representing a different age. Different Groups are measured at one point and compared
        Strengths
        collect data in a short period of time
        → still simulating behavioral changes with age
        Weaknesses
        factors other than age might differentiate groups
        
        cohort effects
        
        generation effects
        
        1 more item...
        
        Longitudinal Design
        within subjects design
        Quasi Experimental
        examines development observing or measuring a group of cohorts over time
        
        "treatment" is age
        
        set of observations -> by a period of development/aging -> set of observations
        
        4 more items...
        
        variables
        
        Quasi independent variable
        (IV): the variable that is used to differentiate the groups of participants or the groups of scores being compared
        
        The nonmanipulated variable
        participant characteristic (such as male, female or timetable)
        
        Statistical Analysis
        
        Pretest-Posttest Nonequivalent Control Group Design
        Statistical analysis of numerical scores
        
        two-factor, mixed design analysis of variance
        
        refer to a statistical software program
        
        Two-Group Designs
        
        compare numerical scores
        
        t-test
        
        single-factor analysis of variance
        
        chi-square for non-numerical data
        
        Factorial Designs
        two or more factors [IV's]
        → show differences within differences
        
        Notation
        
        single factor design
        
        two factor design
        
        Y x Y
        
        presentation in matrix
        
        three factor design
        
        2 x 3 x 2
        
        4 more items...
        
        ... factor design
        
        Factor
        independent variable especially for experiments that include two or more IV's
        
        Advantages
        by more variables
        
        more realistic situation
        
        observable:
        observation of
        
        individual factors
        
        groups factors
        
        Data
        
        Matrix
        
        each cell corresponds to a seperate treatment condition (specific combination of the factors)
        
        The levels of one factor determine the columns and
        the levels of the second factor determine the rows.
        
        provides three seperate
        and distinct sets of info
        
        how factors affect behaviour
        
        2 more items...
        
        Ex.: two factor design
        three seperate sets of mean differences
        
        Mean differences from main effect of Factor A
        
        Mean differences from main effect of factor B
        
        Mean differences from interaction between factors
        
        seperate sets of differences => independent
        
        Main Effect
        = mean differences among levels of one factor
        "There is a main effect for -- such that -- is higher than --"
        
        Two Factor Main Effects
        A two factor study has two main effects; one for each factor
        
        Look for any non-zero differences between the variable-states means
        
        Interaction Between Variables
        the effect of one independent variable depends in the other independent variable
        
        = occurs when two factors, acting together, produce mean differences, not explainable by the main effects
        
        2 more items...
        
        Identify
        → statistical test is needed for significance of interaction
        
        2 more items...
        
        Interpretation
        
        2 more items...
        
        Types of Factorial Design
        
        Mixed Design
        factorial study that combines two different research designs. Common = factorial study with one between-subject factor and one within subject factor
        
        Exp. / Non-+Quasi-Exp. Research Strategies
        one factor = true IV (experimental)
        second factor = quasi-independent variable (non-/quasi-experimental)
        
        2 more items...
        
        Within- / Between-Subject Design
        When the advantage or convenience of a between subjects design apply to one factor, but a within subjects design is preferable for the other factor
        
        2 more items...
        
        Combined Strategy
        uses two different research strategies in same factorial design
        
        Pretest-Posttest Control Group Design
        
        2 more items...
        
        Factors
        
        2 more items...
        
        Higher Order Factorial Designs
        
        more than 3 factors => complex results
        
        results:
        
        main effects for each factor
        
        set of two-way interactions (AxB, BxC, CxA)
        
        three-way-interaction AxBxC
        
        Statistical Analysis of Factorial Designs
        depends on
        
        between-subjects
        
        within subjects
        
        mixture
        
        Compute the mean for each treatment condition (matrix-cell)
        
        Using ANOVA to evaluate statistical significance of mean differences
        
        Application of Factorial Design
        
        Replication of studies
        
        Expansion of studies
        adding a second factor
        → ascertain whether results can be generalized to new situations/populations
        
        Participant Variable as factor in BS Designs
        Reduce Variance Within Groups
        
        No lost external validity
        
        Order of Treatments as factor in WS-Designs
        evaluate order effects
        
        outcomes
        
        3 more items...
        
        Correlational Research Strategy
        two or more variables are measured
        to obtain a set of scores (usually two scores) for each individual.
        
        Strengths an Weaknesses
        
        Strengths
        
        foundation for further directional research
        
        Nonintrusive
        
        High external validity
        
        relationships with more than two variables
        Multiple Regressions
        a statistical procedure for studying multivariate relationships
        Can be used to examine the relationship between two specific variables while controlling the influence of other, potentially confounding variables
        
        Weaknesses
        
        No Information on causality
        
        1 more item...
        
        low internal validity
        
        1 more item...
        
        Goals
        
        describe relationship bewteen two are more variables
        
        no attempt to
        
        manipulate
        
        control
        
        interfere
        
        Differences to
        
        Experimental research
        
        demonstrates cause-and-effect relationship
        
        Differential research
        
        focuses on difference between two groups
        
        Correlational research
        
        intends to demonstrate the existence of a relationship between two variables
        -> no cause-and-effect relationship
        
        Relationship
        
        Numerical Relationships
        interval/ratio
        
        Data presentation
        
        2 more items...
        
        Direction
        
        2 more items...
        
        Form of Relationship
        
        2 more items...
        
        Consistency (Strength)
        
        4 more items...
        
        Non-Numerical Relationships
        ranks
        
        one numerical & one boolean value
        
        2 more items...
        
        Both Non-Numerical Scores from Nominal Scales
        
        2 more items...
        
        Usage of Correlational Research
        
        Prediction
        correlational research allows researchers to use knowledge in a variable to explain (predict)the second variable
        
        criterion variable
        
        2 more items...
        
        Predictor variable
        only predict
        
        2 more items...
        
        Regression
        The statistical process for using one variable to predict the other.
        
        Goal
        
        2 more items...
        
        Reliability
        evaluates the consitency/stability of two measurements
        
        Test-Retest-Reliability
        
        1 more item...
        
        Validity
        evaluates the extent to which the measurement actually measures what it claims to measure
        
        Concurrent Validity
        
        1 more item...
        
        Multiple Regression
        
        for more than 2 variables
        
        used to examine relationship between two specific variables while controlling the influence of other potentially confounding variables
        
        adding predictor variables one at a time shows their individual influence
        
        Descriptive Research Strategy
        
        Goals
        
        describe single variable
        
        describe each individual when several are involved
        
        measures naturally existing variables
        
        capture naturally occurring behavior
        
        Research Designs
        
        Observational Research
        the researcher observes and systematically records the behavior of individuals to describe their behavior
        
        Measurement Problems
        
        2 more items...
        
        Quantify Observations
        
        3 more items...
        
        Sampling Observations
        
        3 more items...
        
        Content Analysis
        techniques of behavioral observation to measure the occurence of specific events in literature, movies, television programs / similar media that present replicas of behaviors
        
        Archival Research
        historical records to measure behaviors or events that occured in the past
        
        Types of Observations
        
        3 more items...
        
        Survey Research
        research study that uses a survey to obtain a description of a particular group of individuals
        Goal:
        obtain a snapshot of the group at a particular time
        Common Application:
        companies seeking accurate descriptions of their customers
        
        Types of Questions
        
        Open-ended-questions
        
        Restricted questions
        
        present a limited number of response alternatives
        
        Rating-scale questions
        
        require selection of numerical value on a predetermined scale
        
        Likert-Scale Rating Scale
        
        1-5 (Strongly Disagree, Disagree, Undecided, Agree, Strongly Agree)
        
        Guidelines to Construct a Survey
        
        demographic questions at the end
        
        sensitive questions in the middle (being warmed up)
        
        group together questions on the same topic (& format)
        
        simple survey format (no overwhelming appearance)
        
        easy to understand vocabulary & language style (target group)
        
        Selecting Relevant / Representative Individuals
        
        select participants to whom the questions are relevant
        
        take time to identify target population
        
        effort to select individuals who accurately represent the group
        
        Need to consult professional help
        
        preparing surveys
        
        specialized recruiting agencies for sampling for large surveys
        
        Administer Survey
        
        internet
        
        mail
        telephone
        
        in person (interviews)
        
        Case Study Research
        
        7. conduct the study
        SESSION 14
        
        decisions on execution of study
        
        Practical Considerations
        
        Before Experiment
        
        plan experiment
        
        rehearse setup & protocol
        
        booking rooms
        
        scheduling recruiting participants
        
        During Research
        
        setting up experiment
        
        reading & signing consent forms
        
        filling out questionnaires
        
        breaks
        
        After Research
        
        saving data
        
        repreparing room
        
        Resources
        
        timeslots
        
        participants
        
        lab space
        
        research assistants
        
        finances
        
        specific participants
        
        Reserach Assistant Instruction
        
        workshop on
        
        setup of experiment
        
        crucial aspects of protocol
        
        behavior in specific situations
        
        interruptions
        
        no-consent
        
        decide which information to share
        
        Participants
        study is a new experience for each participant...
        
        nervous & scared to make mistakes
        
        interrupting/quitting study
        
        showing up to late (recruit more participants than needed)
        
        getting lost in building
        
        do not take study seriously
        
        need clarification or other assistance after study (additionally to debriefing)
        
        Documentation
        thorough & rigorous work needed for data analysis
        Lab notes & documentation of process is important
        
        note everything that can be noticed
        
        disturbances between sessions
        
        questions by participants
        
        quick participants
        
        slow participants
        
        attentiveness
        
        8. evaluate the data
        statistic course
        
        9. report the results
        SESSION 15
        
        Research Report
        includes
        
        statement of purpose
        
        review of relevant background literature
        
        description of relevant methods
        
        summary of results
        
        discussion & interpretation of results
        
        Goals
        provide info on:
        
        what was done
        
        what was found
        
        hot is the study related to the background knowledge
        
        Styles
        
        APA-Style
        
        clear & concise
        
        why research was done
        
        hot research was done
        
        results of research
        
        Writing Style
        
        4 more items...
        
        Submit Manuscript for Publication
        
        select journal that is appropriate for topic if your research report
        
        consult journals "Instructions to Authors" for specific submission requirements
        
        Enclose a cover letter to journal editor along with manuscript
        
        Conference Presentations
        
        submit written summary / abstract to conference organizers for approval
        
        Give oral presentation (10-20 minutes) / present poster (guidelines by organization
        
        Research Proposal
        written report which represents the plan and underlying rationale of a future research study
        
        review of relevant background literature
        
        explanation of how the proposed study is related to other knowledge in the area
        
        description of how the planned research will be conducted
        
        description of possible results
        
        Write a Research Proposal
        
        follow APA guidelines
        
        uses future tense to describe study
        
        abstract is optional
        
        literature review in introduction is more extensive
        
        include combined Results/Discussion // Expected Results section
        
        10. refine of reformulate
        your research idea
        
        ... repeat from step 2
- - - - Theory-Before-Research
        ideas and theories are tested through empirical research
      - Research-Before-Theory
        empirical research as starting point for theory development
      - Spiraling Research Approach
        fluid interaction between theory and data at all stages of the research process
  - - - Historical comparative research,
        examine conditions in historical past and/or in different societies
      - Field Research
        direct observation/recording notes on people in a natural setting for an extended period of time
      - Engagement
        
        Direct
        
        Interviews
        
        face-to-face
        
        telephone interviews
        
        written correspondence
        
        focus groups (guided group conversation)
        
        Structure
        
        structured
        
        semi-structured
        
        unstructured
        
        Ethnographic Field Work
        researcher embeds within a community for a period of time
        
        Action Research
        researcher engages subjects by working towards a shared goal
        
        Unobstrusive data sources
        
        Archival records
        
        news media
        
        actuarial records
        
        official documentary records
        
        autobiographies
        
        private archives
        
        Erosion measures
        
        human traces as data sources
        
        consumption
        
        interactions
        
        frequency of usage
        
        time-data
        
        Accretion measures
        objects/materials used by humans
        
        amount
      - Main Methods
        
        interviews
        
        focus groups
        
        Ethnography
        
        Observations
        
        Content Analysis
        
        Case Studies
        
        (single case/participant studies // for unusual cases)
  - - - Open Coding
        
        first coding
        
        examines data to condense them into preliminary analytic categories or codes
      - Axial Coding
        second coding
        
        organize codes
        
        link codes
        
        discover analytic categories
      - Selective Coding
        last stage
        
        examine previous codes
        
        identify & select data that supports coding categories