Research Methods in Psychology Chapter 8 set

Bivariate Correlation

an association that involves exactly two variables.

What are the steps of describing associations between two quantitative variables?

1. Recording the data

2. Describe the relationship between the two measured variables using scatterplots and the correlation coefficient r. (Or using histograms and t test)

What are Cohen's guidelines for evaluating strengths of association based on r?

.10 is a small or weak effect size

.30 is medium or moderate effect size

.50 is a large or strong effect size

Which graph should be used if both variables are quantitative? And if one or more are categorical?

Quantitative: Scatterplot

Categorical: Historgram bar graph

Which statistics are used if both variables are quantitative? And if one or more variables are categorical?

Quantitative: r

Categorical: t test

What are the two most important validities to interrogate with association claims? Why?

construct validity and statistical validity (third most important would be external validity)

construct validity: An association claim describes the relationship between two measured variables, so it is relevant to ask about the construct validity of each variable. How well were each of the two variables measured?

statistical validity: When you ask about the statistical validity of an association claim, you are asking about factors that might have affected the scatterplot, correlation coefficient r, bar graph, or difference score that led to your association claim. You need to consider the effect size and statistical significance of the relationship, any outliers that might have affected the overall findings, restriction of range, and whether a seemingly zero association might actually be curvilinear.

effect size

describes the strength of an association.

__________ get larger when associations get weaker.

errors of prediction

When all else is equal, a ___ effect size is often considered more important than a ___ one

larger

small

(small effect size can still be significant/important in matters of life and death, such as health research studies)

statistical significance

refers to the conclusion a researcher reaches regarding how likely it is they'd get a correlation of that size just by chance, assuming that there's no correlation in the real world

p value

Helps researchers evaluate the probability that the sample's association came from a population in which the association is zero.

If the p value is very small (less than 5%) we know that the result is very unlikely to have come from a 'zero - association' population and the correlation is considered statistically significant.

____ effect sizes can still be statistically significant if they are identified in a _____ sample

small

very large

outlier

an extreme score - a single case (or several) that stands out far away from the pack

Depending on where the outlier is, it can make a medium-sized correlation appear stronger, or a strong correlation appear weaker than it really is

Outliers matter the most when the sample is ___

small

restriction of range

If there is not a full range of scores on one of the variables in the association, it can make the correlation appear smaller than it really is.

Correction for restriction of range

Formula that estimates the full set of scores based on what we know about an existing, restricted set, then recomputes the correlation.

When can restriction of range apply?

When, for any reason, one of the variables has very little variance.

For example, if researchers were testing the correlation between parental income and child school achievement, they would want to have a sample of parents that included all levels of income. If their sample of parents was entirely upper middle class, there would be restriction of range on parental income, and researchers would underestimate any true correlation.

curvilinear association

The relationship between two variables is not a straight line on a scatterplot, but rather a curve. In this instance, the correlation may seem very small, but in fact is not. (I would suggest not using a scatterplot for this kind of data, but perhaps the histogram)

EX: people's ages and their use of the health care system. This correlation starts with usage very high at a young age, then decreasing 20-50, and then increasing as people reach an elderly age.

What are three differences between causation and association claims?

Causation requires:

1. covariance of cause and effect: here must be correlation, or association, between the cause variable and the effect variable

2. Temporal precedence: The causal variable must precede the effect variable; it must come first in time

3. Internal validity: There must be no plausible alternative explanations for the relationship between two variables.

Directionality problem

When we do not know which variable came first in time (cause and effect)

(temporal precedence)

Third - variable problem

When we can come up with an alternative explanation for the association between two variables, that alternative explanation is the third variable. (internal validity)

To look at the relationship between reaction time and level of expertise in tennis, experts and non-experts are compared. Which of the following would be the most appropriate, easiest way to evaluate the relationship between these variables?

Question 1 options:

correlation coefficient and a scatterplot

Cronbach's alpha and a bar graph

t test and a bar graph

correlation coefficient and a bar graph

...t test and a bar graph

When is an outlier most likely to be problematic?

Question 2 options:

when the sample size is large and the outlier is extreme on both variables

when the sample size is small and the outlier is extreme on one of the variables

when the sample size is small and the outlier is extreme on both variables

when the sample size is large and the outlier is extreme on one of the variables

when the sample size is small and the outlier is extreme on both variables

The Yerkes-Dodson law (1908), shows that performance increases with arousal up to a point, but beyond that, performance decreases with increasing arousal. What type of correlation is this?

Question 3 options:

zero

curvilinear

positive

negative

...curvilinear

Professor Fofana wonders if there is an association between students' grades and whether they complete extra credit in his classes. He makes a scatterplot, with the number of extra credit points earned on the x-axis and the numerical grade in his course without extra credit on the y-axis. He finds that r = 0.28. In addition to this correlation coefficient, what other information would Professor Fofana need to determine if this result is statistically significant?

Question 4 options:

the sample size

the mean of the scores

the effect size

the letter grades of the students

...the effect size

Jenna is interested in the association between the height of professional basketball players and their free-throw shooting percentage. She looks at the correlation between NBA players and their free-throw percentage from last season and she finds a statistically significant negative association. Jenna's friend Elizabeth suggests that Jenna should look at scatter plot of the data. Jenna follows Elizabeth's advice and finds that one of the players is much shorter than the rest of the players and that player has a much better free-throw shooting percentage. When Jenna removes this player from her analysis, she finds that there is no longer a statistically significant relationship between height and free-throw shooting. What kind of problem has Elizabeth helped Jenna identify?

Question 5 options:

a third variable problem

a restriction of range problem

a problem with an outlier in the sample

a moderation problem

...a problem with an outlier in the sample

spurious association

The association is only there because of some third variable

moderator

In association research, when the relationship between two variables changes depending on the level of another variable, that other variable is called a moderator.

Ex: a study looks for moderators in the relationship they find between deep talk and well-being. They wonder if the relationship would differ depending on whether substantive conversations took place on a weekend or weekday (the day of the week becoming the moderator)

the presence of a moderator make hint at ______ _____ ____

weak external validity

Interrogating the statistical validity of an association claim involves five areas of inquiry:

effect size (strength of r)

Statistical significance

the presence of outliers

possible restriction of range

whether the association is curvilinear

The variables in a bivariate correlational study can be either ___ or ___

quantitative or categorical

If both variables are quantitative, the data are usually depicted in a _____; if one variable is categorical, the data are usually depicted in a _____

scatterplot; bar graph

Lack of external validity

1. should disqualify an entire study

2. should not disqualify an entire study

Should not disqualify an entire study. If the study fulfills the other three validities, and its results are sound, the question of generalization can be left for a future investigation.

Suppose that you hear that conscientious people are more likely to get regular health checkups. Which of the following correlations between conscientiousness and getting checkups would probably support this claim?

r = .03

r= .45

r= -.35

r= -1.0

r = .45

Which of these associations will probably be plotted as a bar graph rather than a scatterplot?

The more conscientious people are, the more likely to get regular health checkups

Level of depression is linked to the amount of chocolate people eat

Students at private colleges get higher GPAs than those at public colleges

Level of chronic stomach pain in kids is linked to later anxiety as adults.

Students at private colleges get higher GPAs than those at public colleges.

Which of the following is the essential feature of studies that support association claims?

Question 1 options:

They involve two measured variables.

They involve a correlation between0 and 1.

They involve a correlation between one quantitative variable and one categorical variable.

They involve a correlation between one measured variable and one manipulated variable.

They involve two measured variables.

What other information, in addition to effect size, must you know in order to determine if a correlation is statistically significant?

Question 7 options:

direction of the association

scale of the scatterplot

sample size

external validity

Statistical significance calculations depend not only on the effect size but also on sample size

Which of the following is NOT a way that a researcher might indicate a statistically significant result in a journal article?

Question 15 options:

an asterisk (*)

the word "sig"

a notation of p < .05

a notation of p = .20

a notation of p = .20

Question 19 (1 point) Question 19 Saved

Professor Fofana wonders if there is an association between students' grades and whether they complete extra credit in his classes. He makes a scatterplot, with the number of extra credit points earned on the x-axis and the numerical grade in his course without extra credit on the y-axis. He finds that r = 0.28 and that p<0.001. What does this mean?

Question 19 options:

The result probably came from a zero-association population.

It is very unlikely that this association was found in the sample when, in the full population, there is really no association.

The result is not statistically significant.

There is probably no association between extra credit and course grades in the full population.

It is very unlikely that this association was found in the sample when, in the full population, there is really no association.

Mischel (1972) studied delay of gratification in preschoolers: Children were offered a special reward if they could wait or a less attractive treat if they chose not to wait. A follow-up study was done years later, looking at the same children as adolescents. The researchers found an association between the waiting times of the preschoolers and parents' reports of the same children's behaviors as adolescents. Overall, a positive relation between waiting time as a preschooler and self-control in adolescence emerged. Can a causal relationship be inferred?

Question 24 options:

No, because covariance was not established.

No, because temporal precedence was not established.

No, because internal validity was not established.

Yes; covariance, temporal precedence, and internal validity were established.

No, because internal validity was not established.