Study Guides (400,000)
US (230,000)
UC-Irvine (3,000)
PSYCH (60)
Study Guide

PSYCH 9A- Midterm Exam Guide - Comprehensive Notes for the exam ( 55 pages long!)


Department
Psychology
Course Code
PSYCH 9A
Professor
Bruce Berg
Study Guide
Midterm

This preview shows pages 1-3. to view the full 55 pages of the document.
UC-Irvine
PSYCH 9A
Midterm EXAM
STUDY GUIDE

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

PSYCH 9A - LECTURE 1
RESEARCH METHODS
!
-People do not use appropriate research methods in psychological research run a high risk of generating BS.
-Example: Journal’s Paper on ESP Expected to Prompt Outrage - precognition (ability to predict the future)
for erotic photos.
-The effect size is rather small (53% vs. 50% guessing rate). However, if there are many human research
subjects with each running many trials, such an effect can reach statistical significance as determined by
standard methods of inferential statistics
-If you keep on running experiments of this sort, you will get something that looks like a result of just by
chance.
-The more times an experiment is repeated, the less likely the result is due to chance alone.
-There is no result if it cannot be replicated.
-Basic principles of statistics: The more observations that you make, the more likely it is that you will observe
patterns that “couldn’t possibly occur by chance” - even though they occur just by chance!
-Example: If you keep on running ESP experiments, you are bound to get one that looks as though it is
producing positive results (even though it is just by chance). Proper research methods are needed to interpret
such events.
!
IMPORTANT RESEARCH METHOD: OBSERVATION
-In making the choice about what behavior we should study, we’re usually guided by the questions we hope to
address through our observations. Moreover, we need to formulate the question in a way that leaves no doubt
about how we’re going to link the question to the evidence we collect.
-Induction: on example based on observation of “observed variables”.
-Deduction: apply rules to specific cases.
-Any research projects begins with:
-Testable hypothesis - a specific claim about the facts, framed in a way that will allow an unambiguous test.
(Testability, in turn, is guaranteed by ensuring that the hypothesis is falsifiable).
-Testable hypotheses requires well-defined terms.
-Operation definition - a definition that translates the variable we want to assess into a specific
procedure or measurement.
-We need the operational definition to have construct validity - that is, must truly reflect the variable
named in our hypothesis.
-Example: aggression as an operational action rather than frown - a facial action.
-Untestable hypotheses are ideas that are open-ended or are too general to be interpreted.
-Example: Astrologers predict the future in an open-ended way that we can’t test the claims.
-Scientific method: test a theory by performing an experiment.
*There are many theories out there, based on observation, which do not lead to hypotheses that can be easily
tested by experiment.
-At this time at least, we’re unable to directly manipulate key variables:
-From Physics: Cosmology, original big bang, universe expansion, black hole behavior.
-From Economics: Sunspots cause Economic cycles.
-From Psychology: Darwinian Evolution has played a key role in determining human behavior.
!
!
EXPERIMENTS
-Independent variable(s) - are directly manipulated by the experimenter.
-Dependent variable(s) - are measured.
-Experimental group - the group that is manipulated by the experimenter.
-Control group - the group that is not manipulated by the experimenter.
-Control variable(s) - these are independent variables that are manipulated in an experiment with the aim of
ruling out explanations for the results other than those that bear directly on the tested hypothesis.
-Subject variable(s): theses are characteristics of the (human) subjects who participate in an experiment.
Because they are not directly manipulated by the experimenter, they are not true independent variables.
-Quasi-experiment: an experiment in which only the effects of subject variables are tested (already-existing
groups - groups that experimenter did not create or manipulate).
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

-Example: comparing boys and girls in their level of aggression.
WORKING WITH DATA
* DESCRIPTIVE STATISTICS - to summarize the data.
-Single numbers that each help to describe a distribution of
values include:
-Mean - the average value.
-Median - the middle value.
-Mode - the most frequent value.
-Variability - the measure of the spread of a data set.
- Variance and its square root - standard deviations
(measure of variability about the mean).
-High variability - points deviate from the mean by a lot.
-Low variability - points deviate from the mean by a
little.
* INFERENTIAL STATISTICS - to make inferences based on their data.
-This involves testing a difference between two or more groups of data to see how likely it is that the
difference could have arisen by chance alone.
-Example: Compare human female heights to human male heights.
-Compare the difference between the average height for males and the average height for females to
the spread (variability) in the values found for males and females.
-If the difference between the two averages is large compared to the variation in the measurements,
then the difference is deemed significant. Otherwise, it’s not significant.
-Statistical power: one’s ability to ascertain that a small difference is statistically significant
increases as the number of measurements is increased.
!
CORRELATION
-Correlation - the tendency of two variables to change together (scatter plot).
-Researchers use correlations to summarize the pattern of data.
-Correlation coefficient, r (-1.00 to +1.00).
-Many r values range from 0.40 to 0.60.
-Correlations can be used to:
-Check on reliability of measurements.
-Assess each measure’s validity.
-While there may be a correlation between many variables, there is no reason to think that there is a causal
relationship.
* Correlation is not causation.
!
BETWEEN-SUBJECTS DESIGN
-“Levels” of an independent variable are represented by different groups of subjects.
-Example: clinical drug-testing. Random assignment to “experimental” and “control” groups of subjects.
!
WITHIN-SUBJECTS DESIGN
-A type of experiment design in which all participants are exposed to all different levels of the independent
variable.
-Example: visual psychophysics - each is exposed to all different colors of light and levels of intensity.
!
MIXED DESIGN
-There are at least two independent variables.
-There is at least one independent variable, the effects of which are determined between subjects.
-There is at least one (other) independent variable, the effects of which are studied for single subjects (in all
groups).
!
THIRD OR HIDDEN VARIABLE
-Third-variable problem - the possibility that two correlated variables may be changing together only due to the
operation of a third variable.
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version