After Class Quiz I
If a study is “reliable,” this means that
a. The findings can be generalized to other social settings.
b. The measures devised for concepts are stable on different occasions/raters.
c. The methods are stated clearly enough for the research to be replicated.
d. It was conducted by a reputable researcher who can be trusted.
b. The measures devised for concepts are stable on different occasions/raters.
(remember...the opposite of reliable is "wobbly!" when you think "reliable," think "stable."
It is relatively __________ for any given observation to be within one standard deviation of the mean.
a. unlikely
b. likely
b. likely
(68% of data will fall within 1 SD of the mean!)
Rejecting the null hypothesis is functionally equivalent to concluding that:
a. There IS a difference/association between the variables.
b. There is no difference/association between the variables.
c. Your data are inconclusive.
d. Your results are very likely to happen by chance.
a. There IS a difference/association between the variables.
(remember, null hypothesis = no relationship; if you reject the null, you are saying there is a relationship)
What two factors determine the kind of inferential statistical test used by a researcher?
The two factors that determine the kind of inferential statistical test used are whether you are looking for associations or differences and what kind of variable you are using (nominal, ordinal, interval, or ratio)
You read the manual for a new assessment, and you see that the reliability of the test is very good. You conclude:
a. Nothing—what’s reliability?
b. Nothing yet, until you learn and think more about validity
c. That it is a good assessment to use
d. That you should ask some colleagues about their experiences with the test
b. Nothing yet, until you learn and think more about validity
(remember, you can't have validity without reliability. Having reliability is good, but it doesn't tell you whether or not the test is a truthful representation)
All of the following are common characteristics of experimental research EXCEPT
a. It can produce important knowledge about cause and effect.
b. It relies primarily on the collection of numerical data.
c. It rarely is conducted in a controlled setting or environment.
d. It uses the deductive scientific method.
c. It rarely is conducted in a controlled setting or environment.
(Experimental research is always highly controlled - randomized, controlled trial!)
You are trying to increase the number of socially appropriate comments being used in a 30-minute session. Relative to treatment as usual (i.e., no treatment) control group, Treatment A shows a Cohen’s d of 1.3. What does this tell you?
a. Treatment A has a relatively small clinical effect.
b. Response in Treatment A is more variable than in the treatment group.
c. Treatment A is more effective than treatment as usual.
c. Treatment A is more effective than treatment as usual.
(remember, a Cohen's d at .8 or greater indicates a strong effect size!)
A p-value above .05 can be interpreted to mean that
a. The alternative hypothesis was supported.
b. The results from that particular statistical test are very unlikely to happen by chance.
c. The researchers did not find statistically significant evidence of an association/difference in the variables.
d. The null hypothesis should be rejected.
c. The researchers did not find statistically significant evidence of an association/difference in the variables.
(remember, p<.05 is the minimal standard for showing significance)
What is the difference between an alpha level of .05 and an alpha level of .01? That is, how does it change what you are looking for as a researcher?
An alpha of .05 tells you that there is a 5 in 100 chance that you would've obtained your data given the null hypothesis is true.
An alpha of .01 tells you that there is a 1 in 100 chance that you would've obtained your data given the null hypothesis is true.
The lower/smaller the alpha level, the greater assurance the researcher has that the results obtained are very unlikely to have happened by chance and that they are therefore in support of the alternative hypothesis.
Because entire populations are often difficult to study, researchers often include a subgroup, called a __________, in their research project.
a. clinical population
b. sample
c. analogue
b. sample
Which type of research provides the strongest evidence about the existence of cause-and-effect relationships?
a. Nonexperimental research
b. Experimental research
b. Experimental research
You read a study reporting that autism spectrum disorder (ASD) symptom severity and expressive vocabulary are negatively correlated, r = –.43. Which of the following statements are true?
Select one or more:
a. Children with lower symptom levels tend to have larger vocabularies.
b. Higher symptoms levels cause children to have smaller vocabularies.
c. Children with higher symptom levels tend to have smaller vocabularies.
d. The association between symptoms and vocabulary is weak, according to standards in the field.
a. Children with lower symptom levels tend to have larger vocabularies.
AND
c. Children with higher symptom levels tend to have smaller vocabularies.
(remember, correlations do not imply causation; a correlation of .4 is not weak; a negative correlation implies if variable x does one thing, variable y does the opposite)
A p-value below .05 can be interpreted to mean that:
a. The null hypothesis should be retained.
b. The results from that statistical test are likely to happen by chance.
c. The test result was not statistically significant.
d. The researchers found statistically significant evidence of a difference/association.
d. The researchers found statistically significant evidence of a difference/association.
(remember, p<.05 indicates the data is significant and that therefore you can reject the null and there is a relationship between the two variables)
The interpretation of a p-value varies depending on what test was used to obtain it:
True OR False?
False!
A p-value is always interpreted the same way, regardless of the test statistic.
A single-subject time series employing the AB design is questionably __________.
a. reliable
b. quantitative
c. useful
d. valid
d. valid
(remember, unless you employ ABAB, you cannot be sure that the changes in B are reflective of the treatment - they may simply be due to confounding variables/other changes in the subject)
In __________, random assignment to groups is not used, but the researcher is interested in cause and effect.
a. experimental research
b. quasi-experimental research
c. quantitative research
d. basic research
b. quasi-experimental research
(remember, quasi-experimental research uses a control group and multiple measurements. it does not use randomization or blinding)
Why do confounding variables threaten the validity of a research study?
Confounding variables threaten the results of a research study because they can also affect the dependent variable or outcome, thus making it unclear whether the outcome is due to the independent variable or another unmeasured factor.
The null hypothesis can be described as:
a. The same as the research hypothesis
b. A statement of the expected results
c. A statement of probability
d. A statement that there is no association/difference between variables
d. A statement that there is no association/difference between variables
(remember, think "null = NO" relationship)
Every statistical test is based on null and alternative hypotheses:
True OR False
True!
(remember, an experiment may have multiple hypotheses depending on the variables being measured!)
Name the DEPENDENT variable(s) in this study:
ABSTRACT Thirty-five children who received an autism spectrum diagnosis at the age of 2 years (24 with autism, 11 with PDD-NOS) were re-evaluated 2 years later to examine factors related to the development of spoken language. Child variables (play levels, motor imitation ability and joint attention) and environmental variables (socioeconomic status and hours of speech/language therapy between ages 2 and 3) were used to predict an aggregate measure of language outcome at age 4. After controlling for age 2 language skills, the only significant predictors were motor imitation and number of hours of speech/language therapy. Implications of these results for understanding the early developmental course of autism spectrum disorders and the effects of intervention are discussed.
The DEPENDENT variable is:
Language outcome at age 4
What is the defining characteristic of experimental research?
a. Completion in a research laboratory.
b. Participants are randomly assigned to groups.
c. Cross-sectional data collection.
d. The researchers are interested in quantifying behavior.
b. Participants are randomly assigned to groups.
*RANDOM ASSIGNMENT is the hallmark of experimental research
Please match each measurement strategy with the most appropriate variable (nominal, ordinal, interval, ratio):
1. Diagnosis (ASD, SLI, ID, typically developing)
2. Mother’s education (high school, college degree, or graduate degree)
3. Number of vocabulary items checked off by parent on vocabulary checklist)
4. IQ score (that is, a “standard” score)
1. Diagnosis (ASD, SLI, ID, typically developing) - Nominal (it's a category)
2. Mother’s education (high school, college degree, or graduate degree) - Ordinal (the levels are structured in a hierarchy)
3. Number of vocabulary items checked off by parent on vocabulary checklist) - Ratio (could be any number, including "0")
4. IQ score (that is, a “standard” score) - Interval (there is no "0" IQ)
When we say a result is “statistically significant,” what it means is:
a. Our p-value was above .05
b. It is really unlikely that we got that result by chance (assuming the null hypothesis is true).
c. We proved the alternative hypothesis to be right.
b. It is really unlikely that we got that result by chance (assuming the null hypothesis is true)
(remember, the alternative hypothesis isn't a guarantee of TRUTH, it's just saying that the data in this case gives us evidence to support the alternative over the null)
Why would a larger test statistic generally achieve a higher level of significance?
A larger test statistic will be further away from the mean. The further away from the mean, the higher chance a test statistic will achieve significance at a smaller alpha level.
(remember, the further out on the tail a data point falls, the less likely it is to have obtained it!)
Name the INDEPENDENT variable(s) in this study:
ABSTRACT Thirty-five children who received an autism spectrum diagnosis at the age of 2 years (24 with autism, 11 with PDD-NOS) were re-evaluated 2 years later to examine factors related to the development of spoken language. Child variables (play levels, motor imitation ability and joint attention) and environmental variables (socioeconomic status and hours of speech/language therapy between ages 2 and 3) were used to predict an aggregate measure of language outcome at age 4. After controlling for age 2 language skills, the only significant predictors were motor imitation and number of hours of speech/language therapy. Implications of these results for understanding the early developmental course of autism spectrum disorders and the effects of intervention are discussed.
The INDEPENDENT variables are:
1) child variables (play, motor ability, joint attention)
AND
2) environmental variables (SES, hours of S/L tx)