Term
Conflicts of the Nature of Science:
|
|
Definition
| There have been conflicts b/w religion and science in past and there are still conflicts w/ science in present day (not necessarily w/ religion). Examples: Media (the commercialization of science, advertising, etc), courts and legal system (“expert” testimony, eyewitness, etc), government and politics (global warming controversy, science becoming political.) |
|
|
Term
|
Definition
1. Empiricism (look and see)
2. Control
3. Precision/Accuracy
4. Honesty/Truthfulness
5. Critical/Skeptical
6. Curiosity/openness
7. Parsimony (minimalizing)
8. Abstractness (generalizations)
9. Deterministic
10. Neutrality/Objectivity
11.Publicness
12.Cumulative Enterprise (standing on the shoulders of giants)
13.Rational/Logical/Reasoned
14.Testable/Rigorously Evaluated
15.Practical/Applied (there is a need for both) |
|
|
Term
|
Definition
Observable, look and see attitude of science.
Observations can be distorted. Accuracy, selectivity and bias are all issues. Science probably shouldn’t be based solely on observables. |
|
|
Term
|
Definition
| Used in manipulation of IV’s, randomizing of subjects, double blind experiments, and environmental controls. |
|
|
Term
|
Definition
Validity and reliability. Instrumentation, operational definitions, etc. Limits --- differences in operational definitions. |
|
|
Term
| Operational Definition: Limits?: |
|
Definition
A precise specification of procedures used in an experiment. (of manipulations and measures).
Equivalence of construct and OD, not always in line. |
|
|
Term
|
Definition
| The falsification of data is the “cardinal sin” of science, though there are some prominent examples. Honesty is the assumed procedure. |
|
|
Term
|
Definition
Be skeptical of things that seem out of typical realm of possibility. |
|
|
Term
|
Definition
| This attitude can sometimes conflict with skeptical attitude. Be open to new ideas and willing to study them. |
|
|
Term
|
Definition
| When you find something interesting to study, drop everything and study it. |
|
|
Term
|
Definition
May constrict openness. Don’t want to deviate from the accepted theory. |
|
|
Term
|
Definition
| Occams razor (simplest theory is best). Minimal theorizing. |
|
|
Term
|
Definition
| Generalizations. Law of large numbers, value of group data, etc helps to develop generalizable theories and findings. |
|
|
Term
|
Definition
| Searching for orderly causes and predictability. Everything has a cause. Physical causes for psychological processes. |
|
|
Term
|
Definition
| Desire to minimize bias. Problem with sponsorship; creates vested interests which creates bias. |
|
|
Term
|
Definition
| Scientists should be willing to share their findings and place them up for public scrutiny (peer review, replication, etc) |
|
|
Term
|
Definition
| Allowing peers in the scientific community to review and make suggestions to experiments, design, etc. |
|
|
Term
|
Definition
This is controversial with so many research projects being funded by corporations, people patenting their findings, university research parks, etc. |
|
|
Term
|
Definition
| We should strive to stand on the shoulders of giants. Utilizing literature and theory, meta-analysis, self correcting, etc. |
|
|
Term
|
Definition
| Tied with the skeptical attitude. Hypothesis are stronger when supported/interwoven with other theories or rational explanation. |
|
|
Term
|
Definition
The “show me” attitude. Seeking confirmation. |
|
|
Term
|
Definition
| Research on the basics of a theory or a subject. Not conducted for specific outcomes; done to increase and expand general knowledge. |
|
|
Term
|
Definition
| Research done for specific outcomes. More “practical” in eye of the public. Want immediate benefit. |
|
|
Term
|
Definition
| Sources where studies were first published. Journals, conference reports, presentations, poster sessions, review articles, books, etc. |
|
|
Term
|
Definition
| Psych Bulletin, Psych review, Annual Review of Psychology |
|
|
Term
|
Definition
| A statistical compilation/combination of prior publications. |
|
|
Term
|
Definition
| A form of literature review; going through past articles and recording how many succeeded and how many didn’t. |
|
|
Term
|
Definition
| Standardized measure of treatment effect. The difference between means from different conditions. |
|
|
Term
|
Definition
| Shows how much the treatment helped/harmed. Relates to the standard deviation. (eg: Mean of .5 means the treatment improved/harmed by half a standard deviation.) |
|
|
Term
|
Definition
| People tend to shove studies to the side of the bottom of their file drawers when they fail to find significance, don’t support the hypothesis, or are rejected for publication. (this particularly creates problems for meta-analysis ---a treatment may be seen as more effective simply because studies that DON’T support the treatment don’t get published. |
|
|
Term
|
Definition
Using PsychInfo, find and article and look to see how many times it has been cited. By following up with those articles, we have a better understanding of how the research has progressed, what new things have been found, etc. |
|
|
Term
|
Definition
|
|
Term
|
Definition
| Characteristics of the sample. |
|
|
Term
|
Definition
Characteristics of the population. |
|
|
Term
|
Definition
| All people who participated in a study w/in a population. (Hopefully representative of the population.) |
|
|
Term
|
Definition
|
|
Term
|
Definition
Characteristics of the population. |
|
|
Term
|
Definition
Indicates how variable the data is. (s) |
|
|
Term
|
Definition
| The squared standard deviation. The average of the squared differences from the mean. |
|
|
Term
|
Definition
| Standardized score in standard deviation units (-2, -1, +1, +2, etc) |
|
|
Term
|
Definition
| The prediction if the treatment is not significant. Status quo. |
|
|
Term
|
Definition
Shows us the pattern outcomes if things happen by chance. |
|
|
Term
|
Definition
| Specifies the nature of the sampling distribution. Middle is at zero, variability is caused by standard error, and the shape should typically be normal. |
|
|
Term
|
Definition
The standard deviation of the sampling distribution. |
|
|
Term
| Shape of Sampling Distribution: |
|
Definition
| Should be normal ( a bell curve shape). |
|
|
Term
|
Definition
| A study has significance when the results are unlikely to have happened by chance. |
|
|
Term
|
Definition
There is a 5% chance that the results are due to chance. |
|
|
Term
|
Definition
A standardized score, indicating the location of the studies outcome in the sampling distribution. Found by dividing treatment effect by random error. |
|
|
Term
|
Definition
| Saying that a treatment was significant when it was actually due to error. Probability of committing Type I error is set by the level of alpha. |
|
|
Term
|
Definition
Saying that a treatment was not significant when it actually was. Probability of type II error increases when alpha is too small/power is too small. Probability of type II error decreases when number of participants increases. |
|
|
Term
|
Definition
| Probability level for type I error. |
|
|
Term
|
Definition
| Defines the probability of type II error. |
|
|
Term
|
Definition
| 1 - beta. probability of detecting a significant effect when there actually IS one. When power increases, probability of type II error decreases. |
|
|
Term
|
Definition
| When research is underpowered, the chance of type II error increases, making it more likely that the researchers won’t detect a significant effect, even when one is present. |
|
|
Term
|
Definition
when there is more than one group in a research study. These groups are composed of different participants and are not linked to each other in any way. |
|
|
Term
|
Definition
When multiple measures are conducted using the same group of participants. Ex: A pre-test and a post-test in a study. Advantages:using the same participants eliminates individual differences and decreases error.
|
|
|
Term
|
Definition
Analysis of Variance. Compares multiple means and uses variance estimates to test for significance. |
|
|
Term
| Partitioning of variance: |
|
Definition
| ANOVA finds the total variance and splits it depending on which method is being used. With independent groups, total variance is partitioned between Treatment variance and error variance. In a repeated measures ANOVA the total variance is s[lit by treatment variance, subjects variance, and error variance. In a two way ANOVA, variance is partitioned by the main effect of factor A, Factor B, the interaction of A and B, and error. |
|
|
Term
|
Definition
| The variance among means (between treatment condition and control condition, etc). |
|
|
Term
|
Definition
| The variability with groups (differences between participants, environment, etc). |
|
|
Term
|
Definition
| Treatment variance divided by error variance. There is a sampling distribution for F ratios – skewed to one side (?) |
|
|
Term
|
Definition
| Shows whether a single variable had an effect on its own, apart from other variables. |
|
|
Term
|
Definition
Shows whether multiple variables have an effect when paired together. |
|
|
Term
|
Definition
| Variance between participants in a study. Comes out of the error variance. (repeated measures ANOVA) |
|
|
Term
|
Definition
| Statistical post test analysis that looks for patterns and relationships that would not typically be detected in the original statistical analysis. |
|
|
Term
|
Definition
| When you conduct multiple tests, all with an alpha of p<.05, chances of committing type I error increases dramatically. |
|
|
Term
|
Definition
| Dividing the alpha level by the number of tests conducted. Eg:Two tests = .05/2 = .025. |
|
|
Term
|
Definition
| Post hoc test which separates means to see which means are significantly different from one another. A fairly conservative post hoc test. |
|
|
Term
|
Definition
| The assumption that the variance within groups is equal. (assumption of ANOVA). |
|
|
Term
|
Definition
| Called a one tail test because all of the error is placed at one end of the distribution. |
|
|
Term
|
Definition
| Degree to which a study is actually studying the variables it claims to be studying. |
|
|
Term
|
Definition
Degree to which a study can legitimately cause-effect statements. |
|
|
Term
|
Definition
| Degree to which results can be generalized to different people, settings and times. |
|
|
Term
|
Definition
Characteristics of individuals that can’t be directly observed, such as mental states, traits, abilities, and intentions. |
|
|
Term
|
Definition
| Emphasis on theory with less emphasis on data. These are typically large scale theories. |
|
|
Term
|
Definition
Making sense of tons of data and formulating a small theory. Less emphasis on the theory, more emphasis on the data. (“mini-theories”) |
|
|
Term
|
Definition
| Part of the traditional model of theory constructs. The “feedback” stage after data collection. |
|
|
Term
|
Definition
The hypothesis testing stage of the traditional model of theory constructs. |
|
|
Term
|
Definition
| A variable that intensifies, weakens, or reverses the relationship between the IV and the DV. It alters the relationship between the IV and the DV. |
|
|
Term
|
Definition
| Variables that intervene between the IV and the DV. Could be cognitive, physiological processes, etc that intervene between input and output. |
|
|
Term
|
Definition
| One of the functions of theory. Must be able to demonstrate that operational definitions are representative of broader constructs. |
|
|
