Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
Quant Midterm
terms (2-5) (technically chapter 7 in the book)
| Term | Definition |
|---|---|
| Internal Validity | refers to the extent to which an investigation rules out or makes implausible alternative explanations of the results |
| External Validity | To what extent can the results be generalized or extended to people, settings, times, measures/outcomes, and characteristics other than those included here |
| Construct Validity | Given that the experimental manipulation or intervention was responsible for change, what is the conceptual basis underlying the effect? |
| History | a threat to internal valid. refers to any event that happens to virtually all of the subjects that could explain findings. Often co-occurs with maturation. |
| Maturation | a threat to internal valid. refers to processes within the participants that change over time. Includes: growing older, wiser, and more tired/bored. Only a prob if design cannot separate its effects from the intervention. |
| Testing | a threat to internal valid. refers to the effects that taking a test one time may have on subsequent performance. Control Condition with repeated testing helps. |
| Effect size | refers to the magnitude of the difference between two (or more) conditions or groups and is expressed in standard deviation units. Is equal to the difference between the means, divided by the standard deviation. |
| Instrumentation | a threat to internal valid. refers to changes in the measuring instrument or measurement procedures over time |
| Response shift | a threat to internal valid. refers to changes in a person's internal standards of measurement. Client's threshold has changed, but their symptoms have not (Hello-Goodbye Effect) |
| Statistical Regression | a threat to internal valid. refers to the tendency for extreme scores on any measure to revert (or regress) toward the mean (Average score) of a distribution when the measurement device is re-administered |
| Selection Bias | a threat to internal validity. refers to systematic differences between groups before any experimental manipulation or intervention is presented |
| Attrition | a threat to internal validity. Loss of subjects, occurs when an investigation spans more than one session. Most occurs early on after one or two sessions. Groups no longer randomly comprised. |
| Diffusion or Imitation of Treatment | a threat to internal validity. when intervention is inadvertently provided during times when it should not be or to persons who shouldn't receive it yet. Also when intervention not given when should. |
| Reactions of Controls and Special Treatment | additional threats to internal validity. A fix, receiving treatment as usual. |
| Sample Characteristics | threat to external validity. The extent to which the results can be extended to subjects or clients whose characteristics may differ from those included |
| Narrow Stimulus Sampling | threat to external validity. the extent to which the results might be restricted to a range of sampling of materials (stimuli) or other features (experimenters) used in the experiment. |
| Reactivity of Experimental Arrangements | threat to external validity. refers to being aware of participating and responding differently as a result |
| Reactivity of Assessment | threat to external validity. refers to the extent to which subjects are aware that their behavior is being assessed and that this may influence how they respond. |
| Test Sensitization (pretest sensitization) | threat to external validity. Measurement in the experiment may sensitize subjects to the experimental manipulation so that they are more or less responsive than they would have been had there been no initial assessment. |
| Multiple-Treatment Interference | threat to external validity. When ppl are exposed to more than one treatment. Conclusions restricted, results may only apply to other persons who experience both of the treatments in the same way or order. |
| Novelty Effects | threat to external validity. The possibility that the effects of an manipulation depend upon the newness in the situation. The results attributed to the manipulation may be restricted to the context in which that is novel or new in some way. |
| Generality Across Measures, Setting, and Time | threat to external validity. do these results apply to a bunch of others. |
| Proof of Concept | When we don't care about external validity. to see whether something could occur that is important in principle or theory |
| Priority of internal validity | One must first have an unambiguous finding before one can ask about its generality |
| Problems of Construct Validity | currently pertain to ambiguities about interpreting the results of a study |
| Attention and Contact | threat to construct validity. refers to the extent to which an increase of attention to the client associated w intervention could plausibly explain the effects attributed to the intervention. |
| Single Operations and Narrow Stimulus Sampling | threat to construct validity. when a single set of stimuli, investigator, or other facet of the study that is considered irrelevant may contribute to the impact of the IV. |
| Experimenter Expectancies | threat to construct validity. Unintentional effects the experimenter may have that influence the subject's responses in the experiment. |
| Demand Characteristics | threat to construct validity. Cues of the experimental situation that are ancillary to what is being studied but may provide info that prompts behavior that could be mistaken as a treatment effect. |
| Data-Evaluation Validity | refers to those facets of the statistical evaluation that influence the conclusions we reach about the experimental condition and its effect. |
| Alpha (Type I error) | probability of rejecting the null hypothesis when its true. the independent construct did not cause or is related to the dependent construct. (false positive) |
| Beta (Type II error) | probability of accepting the null hypothesis when it is false. concluding that the difference b/t two groups is random, when not. (Danny Phantom) |
| Power (1 - B) | probability of rejecting the null hypothesis when IT IS false or the likelihood of finding differences between conditions when, in fact, the conditions ARE truly different. |
| Effect size | a measure of the size of the difference between two groups in standard deviation terms. A function of the variability found in the samples examined as well as the sample size. Large this = more Power. |
| Standard Deviation | a measure of variation or variability about a mean. Understood also as the square root of the variance. |
| Low Statistical Power | threat to data eval valid; is when the ___ is too low to detect an experimental effect. Can't say if truly no effect, or just low ___. |
| Subject Heterogeneity | threat to data eval valid. When the sample is too different, increases variability (denominator in the effect size formula). Decreasing heterogeneity will lower external validity. |
| Variability in the Procedures | threat to data eval valid. variation in _____ leads to increases in variability -> reductions in effect size and power. |
| Unreliability of the Measures | threat to data eval valid. Error in the measurement procedures introduces obscuring validity. More reliable measures, more power, less variability. |
| Restricted Range of Measures | threat to data eval valid. if limits in what a measure can pick up, or have floor or ceiling effects (cannot catch below or above a certain level) changes due to IV may not be detected |
| Errors in data recording, analysis, and reporting | threat to data eval valid. inaccuracies in recording, systematic and unsystematic errors. |
| Systematic Errors | means that scores or characteristics of the subjects were miscoded or recorded in the same direction. May alter affirmative conclusions. |
| Unsystematic Errors | means that errors were random or showed no pattern. May negate or obscure group differences because the errors add variability to the data. |
| Multiple Comparisons and Error Rates | threat to data eval validity. when multiple stats tests are completed within same investigation, likelihood of a "chance" finding increases. False conclusions more likely unless p is adjusted |
| Misreading or Misinterpreting the Data Analysis | threat to data eval validity. The conclusion reached from the data analysis are not sanctioned by the data/statistics. Either proper statistic not run or conclusion extends beyond statistic scope. |
| Risk Factor | a predictor of some later outcome. it is a correlation where we know that one variable comes before the other. Correlation where timeline is clear |
| Protective Factor | a variable that prevents or reduces the likelihood of a deleterious outcome. Negatively correlated with the onset of some later problem. |
| Moderator | A characteristic that influences the direction or magnitude of the relationship between an independent and dependent variable. will a particularly finding generalize to all subjects |
| Mediator | An intervening variable that may account (statistically) for the relationship between IV and DV. Points to a possible mechanism but is not necessarily a mechanism. |
| When Mediation is Evident | Intervention leads to change in outcome measure. Intervention alters the proposed mediator. The mediator is related to outcome. Outcome effects not evident or substantially less evident when mediator did not change. |
| Mechanism | refers to a greater level of specificity than mediator and reflects the steps or processes through which the intervention actually unfolds and produces the change. |
| Moderated Mediation | occurs when the strength (or direction) of the relation of the mediator to outcome depends on the level of some other variable. |
| More on Moderated Mediation | A given outcome can be reached through different means (mediators), a single outcome may have many paths, and these paths may reflect different mechanisms, which depend on other variables (moderators), leading to an outcome. |
| Case-Control Designs | Observe two or more groups who vary on a specified characteristic at a given point in time. Goal is to measure whether the groups who vary on this characteristic also vary on other characteristics. Looks at DV, search for possible IVs associated with DV |
| Matching | Allows the investigator to match (equalize) subjects on one of the variables assessed at pretest that may influence the results |
| Cross-Sectional Design | Type of case control design. Goal is to examine factors that are associated with particular characteristic of interest at the current point in time. Useful for identifying correlates and associated features. |
| Retrospective Cross-Sectional Design | Type of case control design. Tries to determine if any past events were associated with a DV. |
| Advantages of Cross-Sectional studies | Efficient way to collect info about DX categories. Feasible and efficient in terms of costs and resources. No worries about attrition. Make it easy to study moderator variables. |
| Disadvantages of Cross-Sectional studies | Findings are correlational. Selective responsiveness (e.g., choosing ppl who chose to participate and comparing them to those who didn't). Memory is not reliable. Accuracy in creating comparison groups (e.g., reliability of dep DX to classify) |
| Cohort Designs (Longitudinal or Prospective Studies) | Seek out antecedents that may be risk or protective variables of a dependent variable. Group is followed over time. Strength is establishing temporal time line and relations -> important in establishing cause and correlation |
| Single-Group Cohort Design | type of cohort design. selects a single group and examines selected characteristics over time. |
| Birth-Cohort Design | type of cohort design. creating groups (cohorts) based on birth year (e.g., 2000-2005) and follow over time. Limited generalization outside of those born in that cohort. |
| Multicohort Design | type of cohort design. Multiple version of single-group design. Groups to be compared differ on a characteristic. |
| Accelerated Multi-Cohort Longitudinal Design | type of cohort design. A long period of time is staggered to study a long period in a shorter time. EX: Study of 5-14. Follow three groups: 5-8; 8-11; 11-14. Measure at same start and intervals. |
| Strength of cohort designs | Permit within group (within single cohorts) and between group (in multicohort studies) comparisons of temporal associations of IV and DV. One can establish trajectory of events that lead to a particular outcome. Can help establish risk + protective factor |
| Weaknesses of cohort designs | Expensive and challenges to carry out. Attrition is problem, hard to track and keep participants motivated. Long time, update to instruments or add new ones (instrumentation). Changes in researchers. May not tell you what causes what. |
| Sensitivity | refers to the rate or probability of identifying individuals who are predicted to show a particular characteristic (based on a measure/variables) and DO show that outcome. True positives and true negatives. |
| Specificity | refers to the rate or probability of identifying individuals who are not likely to show an outcome and DO NOT. |
| Predicting DVs from previously correlated variables (particularly risk/protective) | Calculate measures of sensitivity and specificity. Need to decide the optimal rates of both to make predictions for individuals or groups. "Optimal" depends on costs/benefits of correct (true positive/negative) versus incorrect (false positive/negative) |
Created by:
gracearmstrong
Popular Math sets