Chapter 14- Generalizing Results
Smart (1966)- college students were studied in over70% of the articles published between
1962 and 1964. Unrepresentative subjects, yet easily attainable.
Researchers use either males or females simply because this is convenient or the
procedures seem better suited to either males or females.
Gender bias may arise: confounding gender with age or job status and selecting response
measures that are gender-stereotyped.
Solution= be aware of possible gender differences and include both males and females in
Participants in one locale may differ from participants in another locale.
Findings obtained in one geographic region may not generalize to people in other regions.
Statistical Interaction (Generalization):
Generalization is like an interaction in a factorial design- this interaction occurs when a
relationship between variables exists under one condition but not under another.
Researchers can address generalizability by including the subject types (gender, age,
ethnicity) as a variable in the study.
Criticism does not mean that results cannot be generalized.
Replication of studies acts as a safe guard against limited generalizability.
More recently, most samples of college students are ethnically diverse
External validity of the research is increased
Cultural research attempts to identify similarities and differences that may exist in
personality and other psychological characteristics as well as ways that individuals from
different cultures respond to the same environments.
Broader view of the importance of cultural diversity needed (Miller 1999)
Generalizing to other Experimenters:
Conductor of the study is the source of a generalization issue-must make sure any
influence the experimenter has on the subjects is constant through entirety of the
Characteristics: Experimenters personality, gender, the amount of practice in the role of
experimenter. Ex: participants perform better by experimenters of the opposite sex.
Solution: use two or more experimenters, both male and female, to conduct the study
www.notesolution.com Pretest Generalization:
Limits the ability to generalize to populations that did not receive a pretest, in the real
world people are rarely given a pretest.
Importance: allows researchers to assess mortality (drop outs) effects when it is likely that
some participants will withdraw from the experiment- ability to tell whether the people
who withdrew are different from those who completed the study.
Solution: Solomon four-group design: half receive a pretest and half receive only the post-
Laboratory Setting Generalization:
Allows experimenters to study the impact of I.V under highly controlled conditions
Whereas, in a field study the experimenter examines and manipulates the I.V in a natural
Solution: conduct experiment under both a field study as well as in a laboratory setting
Replication- overcoming any problems of generalization that occur in a single study
a) Exact replicationattempts to replicate precisely the procedures of a study to see
whether the same results are obtained. Used to build on the findings of a previous study.
Mozart Effect- failure to replicate a study
b) Conceptual replicationattempts to use different procedures to replicate a research
finding. The same I.V is manipulated in a different way. Does the relationship hold when
other ways of manipulating or measuring the variable are studied (generalization).
Meta-Analysis- researcher combines the actual results of a number of studies. Focuses on effect
size, table will show the effect size obtained in a number of studies along with a summary of the
average effect size. Allows statistical, quantitative conclusions.
Literature Review (1)- summarizes what has been found (2)-tells the reader what findings are
strongly supported and what are weakest (3)- points out inconsistent findings and areas where
research is lacking (4)- discusses future directions for research. Allows for the possibility of future
direction in studies and identifies trends in literature.
External Validity- degree to which the results of an experiment may be generalized
www.notesolution.com Chapter 12- Understanding Research Results (Description and Correlation)
Scales of Measurement:
Mainly all Independent Variables are nominal, right and left-handed individuals are an
Whenever a variable is studied- there is an operational definition of the variable and there
must be two or more levels of variable
Four scales of measurement: (1) Nominal, (2) Ordinal, (3) Interval, (4) Ratio
Nominal = scale of measurement, categories that have no numerical value
Ordinal = measurement categories form a rank along a continuum
Interval = intervals between numbers on the scale are all equal in size
Ratio = absolute zero present, (physical measures such as weight, or timed measures such
as duration or reaction time)
Analyzing the Results:
Comparing group percentages
Correlating individual scores
Comparing group means- compare the mean responses of participants in two or more
Indicates the number of individuals that receive each possible score on the variable
Pie charts- divide a whole circle into slices that represent relative percentages. Good for
nominal scaled data.
Bar graphs- separate and distinct bar for each piece of information. (x= possible
responses) (Y=number of individuals who chose each response)
Frequency polygons- line to represent frequencies.
Histograms- bars to display frequency distribution for a quantitative variable. The values
are continuous. Depicts what scores are continuous and most frequent.
Allows researchers to make precise statements about the data. Two numbers: 1 number
that describes how participants scored overall, 1 number describes the variability
Mean- obtained by adding all the scores and dividing by the number of scores, overall set
Median- score that divides the group in half
Mode- the most frequent score
www.notesolution.com Standard deviation- indicates the average deviation of scores from the mean
Range- simply the difference between the highest score and the lowest score
Statistic that describes how strongly variables are related to one another
Pearson product-moment correlation coefficient- (r) 0.00 to +/- 1.00. Closer to +/-1.00
indicates a stronger relation
Curvilinear relationship- correlation coefficient will not indicate a curvilinear
relationship as it is only used to depict straight lines. Thus, a scatterplot is required, as the
[r] value would be represented as 0.00.
Refers to the strength association between variables
Calculations used to predict a persons score on one variable when that persons score on
another variable is already known
Y(score to be predicted) = a (constant) + b (weighting adjustment factor) X (slope of the
Used to combine a number of predictor variables to increase the accuracy of prediction of
a given criterion or outcome variable
Correlation between predictor variable and single criterion variable