click below
click below
Normal Size Small Size show me how
201 RMIP
Chapter 6
Term | Definition |
---|---|
Validity | An indication of accuracy in terms of the extent to which a research conclusion corresponds with reality (that is true or correct). |
What are the 4 types of validity? | 1.Internal validity 2.Construct validity 3. External validity 4.Statistical conclusion validity |
Internal valifity | Extent to which a study provides evidence of of a cause-effect relationship between the independent and dependent variables. |
What is the biggest threat to internal validity? | Confounding |
Confounding | Error that occurs when the effects of two variables in an experiment cannot be seperated, resulting in a confused interpretation of the results. |
What is the most funademntal type of validty? | Internal validity |
When some condition co-varies with the independent variable in such a way that their seperate effects cannot be sorted out, the two variables are ______. | Confounded |
Subject variable | A difference between subjects that cannot be controlled but can only be selected (ex: gender, age, poverty, IQ) |
When is the problem of confounding particularly acute in research? | When the experimenter cannot control the independent variable (subject variable). |
Construct validity | Extent to which the results support the theory behind the research . |
What way can you improve construct validity in your research? | Manipulation check |
Manipulation check | Aspect of an experiment designed to make certain that variables have changed in the way that was intended. |
Give an example of a manipulation check (anxiety and learning). | If your experiment is measuring anxiety and how it affects learning, don't automatically assume that fingernail biting is correlated with anxiety. You could do the Beck Anxiety Inventory to make sure that it is. |
External validity | How well the findings of an experiment generalize to other situations or populations (ex: how inaccurate studies can be from the 1950s to now) > concerns whether the results of an experiment would generalize to another time, place, or set of participants |
Ecological validity | Extent to which an experimental situation mimics a real-world situation (laboratory settings vs natural field setting). Closely related but not identical to external validity. |
Statistical conlcusion validity | Extent to which data is shown to be the result of cause-effect relationships rather than accident. Similar to internal validity. |
What question should you ask in regard to checking the statistical conclusion validity of your research? | Did the independent variable truly cause a change in the dependent variables, or was the result accidental, and thus caused by pure chance? |
What are the 7 threats to internal validity? | 1. Ambiguous temporal precedence 2. History 3. Maturation 4.Effect of repeat testing 5.Regression effect 6.Selection 7.Mortality |
By attempting to fix any threats to internal validity, you are... | Learning to avoid the confounding of potentially important variables with the independent vairable or variables of interest. |
Explain ambiguous temporal evidence (as a threat to internal validity). | Although two variables are related, it is unclear which one is the cause and which one is the effect. |
Explain history (as a threat to internal validity). | Events that occur outside of the experiment that could influence the results of the experiment (ex: sunny day vs rainy day on depression studies) |
Explain maturation (as a threat to internal validity). | A source of error in an experiment related to the amount of time between measurements . |
Explain effect of repeat testing (as a threat to internal validity). | Performance on a second test is influenced by simply having taken a first test (generally perform better on the second and later tests than the first). |
Explain regression effect (as a threat to internal validity). | Tendency of subjects with extreme scores on a first measure to score closer to the mean on a second testing, |
Random error (as part of the regression effect, a threat to internal validity) | The part of the value of a variable that can be attributed to chance (ex: luck on a multiple choice test). |
Explain selection (as a threat to internal validity). | A confound that can occur due to assignment of subjects to groups |
Explain mortality (as a threat to internal validity). | The dropping out of some subjects before an experiment is completed, causing a threat to validity. (ex: experiment on diet for weight loss, people who drop out may be more unmotivated than those who stick in it. Invalid results). |
What are the 2 threats to construct validity? | 1.Loose connection between theory and method 2.Ambiguous effect of independent variables |
Explain loose connection between theory and method (as a threat to construct validity). | Poor operational definition of theoretical concept |
Explain ambiguous effect of independent variables (as a threat to construct validity). | Some participants see the situation in the same way as the experimenter, but others understand it differently>experimental circumstances are ambiguous. Or, whenever people are aware that they ae participating in research, they may change their behaviour. |
What are the 2 threats to external validity? | 1.Other subjects 2.Other times |
Explain other subjects (as a threat to external validity). | Participants should be chosen with equal attention to their representativeness relative to some larger population (ex: university students are often participants but do not often represent poor communities). |
Explain other times (as a threat to external validity). | Would the same experiment conducted at another time produce the same results? Many historical trends render particular findings invalid, whether they concern use of language, attitudes towards foreign countries, or perception of deviant groups. |
What is the threat to statistical conclusion validity? | Improper use of statistics in analyzing the data |
What are 2 ways of improperly using statistics in analyzing the data (as a threat to statistical conlcusion validity) | 1.The problem of power 2.Inacurrate effect size estimation |
Explain the problem of power (as a threat to statistical conclusion validity) | The major threat may be the conclusion that the independent variable had no effect, but if your study employed too few subjects or made too few observations, your conclusion may be incorrect. |
Explain inacurrate effect size estimation (as a threat to statistical conclusion validity). | The size of the relationship is measured poorly (you might believe that the independent variable caused a much smaller effect than it actually did, and your conclusions about your study would be incorrect. |
Role demands | Participants' expectations of what an experiment requires them to do. |
What are 2 threats to research validity as a result of role demands? | 1. Good subject tendency 2. Evaluation apprehension |
Good subject tendency | Tendency of experimental participants to act according to what they think the expermienter wants. |
Evaluation apprehension | Tendency of experimental participants to alter their behaviour to appear as socially desirable as possible. |
What is an issue with blind experimenters? | Blind experimenters may devise their own hypotheses about experiments and thus unintentionally bias the participants in the direction of their concocted hypotheses. |
What is a solution to the issue with blind experimenters? | 1. Standardize or automate experiments as much as possible (ex: online questionnaire) |
What is a solution to the fact that experimenters can unintentionally bias the results of an experiment? | Blind experimenters (they are unaware of the purpose of the study) |
Internal validity vs construct validity: what must you rule out from each? | Internal validity: strive to rule out alternative variables as potential causes of the behaviour of interest. Construct validity: Must rule out other possible theoretical explanations of the results. |
Internal validity vs construct validity: how do you fix errors in the research? | internal validity: you can find it possible to redesign the study to control for the source of confounding. construct validity: must deisgn a new study that will permit a choice between the two competing theoretical explanations of the results. |
What are the 4 types of research validity? | 1.Internal validity 2.Construct validity 3.External validity 4.Statistical conlcusion validity |
A research has _____ validity if a cause-effect relationship actually exists between the independent and dependent variables | Internal |
_____ occurs when the effects of two independent variables in an experiment cannot be seperately evaluated. | Confounding |
_____ validity concerns the question of whether the results support the theory behind the research. | Construct |
______ validity concerns whther the results of the research can be generalized to another situation: different subjects, settings, times, and so forth | External |
______ validity concerns the proper usage of statistics. It addresses whether the observed relationship is a true cause-effect relationaship or is accidental. | statistical conclusion |
What are the 7 threats to internal validity? | 1.Ambiguous temporal precedence 2.Events outside the laboratory 3.Maturation 4.Effects of testing 5.Regression effect 6.Selection 7.Mortality |
The ____ ____ occurs when subjects are tested on related measures and there is error in the measurement. Individuals who performed at the extremes on one measure will tend to score closer to the mean on the other. | Regression effect |
What are the 2 threats to construct validity? | 1.Loose connection between theory and experiment 2.Abiguous effect of independent variables |
Among problems that cause an ambiguous effect of the independent variables is the tendency for participants to... | ... interpret conditions differently from the experimenter. |
What are the 3 threats to external validity? | Problems arising from 1.other times 2.other subjects 3.other settings |
What are the 2 threats to statistical conclusion validity? | 1.lack of power 2.inaccurate estimation of size effect |
What are 4 ways of preventing role demands from biasing experimental results? | 1.Invent a cover story>deceives pt about purpose of experiment 2.Divide experiment>part of data collected in another setting 3.Use measures unlikely to be influenced by the pt's expectations 4.Keep pt unaware that experiment is being conducted |
What are the 2 issues as an outcome of role demands when conducting research? | 1.Good-subject tendency 2.Evaluation apprehension |
In what 2 ways can experimenter bias be reduced? | 1.Keep the experimenter from knowing the conditions in the experiment or its purpose 2.Standardize the procedure as much as possible (ex: online questionnaire) |