Transcript Review Unit 2-research -2014-15
Chapter 2: The Research
Understanding and Prediction
Hypothesis
=A tentative statement about two or more variables-a tentative statement about how things work (Students that eat breakfast perform better in school-what are the two variables here?) -
an educated guess
.
RESEARCH
Applied Research
=has clear, practical application; goal to control positive/negative situations • If it is found that students who eat breakfast perform better, schools will initiate a breakfast program
Basic Research =
questions of interest that may not have immediate, real world application • How does anxiety affect people’s desire to be with others (affiliation need)?
• Do different cultures react differently to stress?
THEORY
•
• If we observe there is a relationship between breakfast eaters and performance we formulate a theory
Theory
=
predicts behavior or events
-
only change as new information available
-
more permanent-have considerable facts to support it.
Unit 2 Review
longitudinal study
study same people over a long period disadvantage=expensive; drop out advantage=same cohort, less confounding variables
cross sectional study
different cohorts at the same time, less expensive and less time; disadvantage=different cohorts
Survey-
questionnaire (can study more people, study things not ethical through experiment)
case study-
in depth study of an individual (used for rare occurrences of events/illnesses
experimental study
-manipulation of Independent variable (experimental group receives the treatment/control group does not) to see it’s effect on the dependent variable
•
Experimental Research: Looking for Causes Experiment
= manipulation of one variable under controlled conditions so that resulting changes in another variable can be observed (feeding one group of students) –
Detection of cause-and-effect relationships Variable
=
any measurable conditions, controlled or observed in a study
• •
Independent variable (IV)
see its effect on = variable
manipulated
(food) to
Dependent variable (DV)
= variable
measured
manipulation of IV (school performance) – How does IV (food) affect DV(performance)? and affected by
Operational definitions
precisely define each variable(IV-breakfast, DV-school performance)-
required for good experiment
most needed aspect of a study-
so study can be
refuted
or
verified through REPLICATIOIN
(makes study scientific)
Independent and Dependent Variables Hypotheses
1. Riding the bus to school (IV) makes students more intelligent(DV) 2. Kids who view aggressive cartoons(IV) are more likely to act aggressively(DV) 3. AP Psychology students who eat chocolate(IV) perform better on vocabulary tests (DV)
Operational Definitions=clearly defining independent/independent variables for replication
Children (Male/female, ages 4 – 6) who
view aggressive cartoons
= , viewing all of Sponge Bob, episode 5, while alone in…..
are more likely to
act aggressively
= placed on the playground for 30 minutes with 10 children who did not view cartoon, five minutes after cartoon was shown and strikes another child
•
Experimental and Control Groups: The Logic of the Scientific Method Experimental group
(exposed to manipulation of independent variable-chocolate given) •
Control group
(similar subjects but does not receive IV manipulation given to the experimental group-no chocolate given) • EVERYTHING ELSE FOR THESE TWO GROUPS MUST BE THE SAME
and
Resulting differences in the two groups
must
be due to the independent variable •
Extraneous and confounding variables Levels of independent variable =
1.chocolate verses not getting chocolate; 2. also two independent variables=chocolate and breakfast verses none for Control Group
The Scientific Method: Terminology
Population
=animals or people from which a sample is drawn (all AP students in Broward) and researchers want to generalize about
Participants
or
subjects
=organisms whose behavior observed in a study
Sample
=subjects from the population (AP psych students selected from all schools in Broward County)
Random sampling
-all in population have equal chance of selection.
Representative Sample
is
only way to generalize results to population Random assignment
participants have equal chance of placement in control or experimental groups;
lessons confounding variables
Figure 2.16 The relationship between the population and the sample
Unit 2 Review
Descriptive statistics describes data –
mean, mode , median, standard deviation
Measures of central tendency
= typical or average score in a distribution • •
Mean
: arithmetic average of scores
Median
: score falling in the exact center, or the average of the two center scores •
Mode
: most frequently occurring score mean is most useful measure of central tendency except when
Outliers
= mean distorted by extreme scores
statistical inference-
conclusions drawn about the relationship between variables, from a sample to entire population
Figure 2.11 Measures of central tendency
Descriptive Statistics: Correlation
• When two variables are related to each other, they are
correlated
.
•
Correlation Coefficient
= relationship between two variables • How well does A predict B?
– Strength of the correlation -1.0 to +1.0
positive correlation
-as one variable increases, so does the other ; as one variable decreases, so does the other
negative correlation-
one variable increases the other decreases
Figure 2.14 Interpreting correlation coefficients
Correlation
Correlation
Correlation
Correlation
Correlation
Correlation
Scatter plot =best way to show relationship between variables
3 4 5 6 Hours Spent Watching Television per Day & GPA
PERSON HRS GPA
1 2 0.5 1 3.50
3.75
2 2.5 3 3.5 4.00
2.75
2.75
1.75
7 8 9 10 4.5 5 5 7 2.25
1.50
2.50
1.00
Which statistic approximates the relationship between the variables?
50% N=20 N=10 r=
-
.
90 r=.50
Unit 2 Review
In
normal curve
-
distribution of scores
-68% fall within 1 SD above/below the mean
percentile scores
takers – the same as or better than 72% of population/test
Describing Data
Measures of Variability-how scores vary from the center
• Normal Curve (bell shaped)
Descriptive Statistics: Variability
=
how much scores
vary
from each other
and
from mean
(see the normal curve or bell curve pp 63-64 or 66-Barrons) –
Standard deviation
= how far scores are from the mean/average; for the Normal curve with IQ, one standard deviation is 15 points from the mean If scores deviate 10 points, curve by 10 points, how far will scores deviate from mean???
In normal curve
-
distribution of scores
-68% fall within 1 SD above/below the mean
-
Range
= distance between highest and lowest scores in data set
Figure 2.12 Variability and the standard deviation
Z scores
measure the distance of a score from the mean (either - or +); a z score of -1 is 15 points below the mean, -2 is 30 points below mean
Percentile scores – the same as or better than 72% of population/test takers 38 th percentile =
you did the same or better than 38 percent of people who took a test
Distribution of Scores-Bell Curve
• • •
Symmetrical distribution -see
p. 65 in Barron’s
Positively skewed
(curve to
Left
)-more low scores than high
Negatively skewed
(curve to
Right
)-more high scores than low If Negative skewed due to test scores, can only assume????
Experimental Research:
•
Replication=
repeat a study to see if earlier results are duplicated (this is why the
operational definitions
are important) •
Reliable
when you can replicate or repeat it •
Valid
when it measures what the researcher set out to measure
Statistics and Research: Statistics Drawing Conclusions
– using mathematics to organize, summarize, and interpret numerical data –
Descriptive statistics
(the numbers) organizing and summarizing data (measures of central tendency, measures of variability, and the correlation coefficient)
to see if there is a relationship between variables
–
Inferential statistics
: interpreting data and drawing conclusions about the larger population
Statistical significance
= the relationship found between the IV and DV is not due to chance
(.05 level of significance
)=
less than 5 chances in 100
. It can never be 0 because we can never be 100% certain
Correlation Predicts Strength or Relationship Between Variables DOES NOT Say ONE CAUSES OTHER:
Correlation does not =Causation
– Foot size and vocabulary positively correlated – larger feet belong to older children
Strengths and Weaknesses of Experimental Research
• Strengths: –
conclusions about cause-and-effect
relationships can be drawn • Weaknesses: – artificial nature of experiments – ethical and practical issues
Figure 2.10 Comparison of major research methods
Advantages/Disadvantages of these (naturalistic observation, case studies, surveys) called Descriptive/ Correlation Methods : Advantage:
• explore questions that can not be examined with experimental methods (poor maternal nutrition and birth defects)
Disadvantage
:
Cannot control events
to isolate
cause and effect
Evaluating Research: Methodological Pitfalls
• • •
Sampling bias =
sample not representative of population- I CANNOT DRAW CONCLUSIONS
Placebo effects = participants’ expectations
lead them to experience some change (do better on a test, less headaches), regardless of the Independent Variable/Treatment
Placebo method
helps=give both groups a drug (one real and one a placebo)
Evaluating Research: Methodological Pitfalls
Distortions in self-report data
(survey, interview): –
Social desirability bias =
give socially approved answers to personal questions –
Response set =
respond to questions in a particular way that is unrelated to the content of the question (agreeing with almost everything on a questionnaire)
-Hawthorne Effect
=
changes
in subjects
behavior due to
the
attention of researcher
(having control and experimental groups help)
Evaluating Research: Methodological Pitfalls
•
Experimenter bias
= researchers expectations about outcome of study influences results; treats experimental and control groups differently to increase chance of confirming hypothesis
double-blind control/procedure
= neither subject or experimenter know which group is the control or experimental group
Single blind control/procedure =
the subject does not know if they are the control or experimental group
Ethics in Psychological Research: Do the Ends Justify the Means?
Ethical standards for research: the
American Psychological Association
academic research and the
IRB : ensures ethical treatment for animal and human research :
1. Informed consent
– participant’s permission, told potential risks, offered alternative activity
2. No harm to humans Psychological or physical 3. Minimal harm to animals Ethical Treatment 4. Debriefing to offset deception 5. Confidentiality-
cannot share names (includes test scores-UNLESS WRITTEN PERMISSION PROVIDE
Deception IS PART OF RESEARCH!!!
Figure 2.17 Ethics in research