NSCI 3011 Review

Which of the following research areas is clear on how associations will hold up in the "real world"

Study 1: r = 0.9, P = 0.01

To predict a binary outcome variable from one or more predictor variables

A statistical test to compare differences between non-normally distributerd continuous variables

Mean difference (95% Confidence interval): -0.16 (-0.59, 0.28) implies no evidence of an association

Participants with the disease are first identified and compared to another set who do not have the disease

To test if there is a difference in BMI score between 3 treatment groups

Conclusions on systemmatic reviews & meta analysis consider the quality and size of studies

Reliability = the extent to which a measurement tool gives consistent result

A correlation of zero means_______

T-test _____________

Independent t-test: Compares 2 un-related groups

Select the odd item (....think CRAP test for evaluating scientific articles)

_________ describes exactly how the data were collected

Involves the non-random manipulation of independent variables. A control group may or may not be present.

A ______is exactly what the researchers measured

The extent to which a tool measures what it is supposed to measure

A negative correlation coefficient means ________

Validity = the extent to which a measurement tool gives consistent result

Odds Ratio > 1 means increased odds of the outcome occurring

A statistical test to compare differences between two categorical variables

Correlation coefficient ranges from 0 to 1

Independent and dependent variables are measured simultaneously. Data collected at a single time point

Probabilistic sampling is _______

_________ the big picture or idea that the researcher wants to study

P-value: Probability that results occurred by chance rather than indicating a true association between variables

A correlation coefficient can tell whether an association is positive or negative

Odds Ratio = 1.13 (0.94-1.31)

Mean difference (95% Confidence intervals): [0.91 (0.33, 1.49) implies no evidence of an association

Which of the following is NOT likely to be asked by peer reviewers during a review

A statistical test to compare differences between normally distributerd continuous variables

Which of the following areas of research is most basic (clear on mechanisms)

A statistical test to compare 2 groups on a continuous outcome

P-value = 0.07

Relative Risk for developing CVD with intake of whole grains vs refined grains = 0.80

Inferential statistics: summary of the characteristics of your sample

A________ is the list that the sample will be drawn from

What questions would the IRB likely NOT have for a research protocol involving children?

This type of plot gives us information about publication bias

Systematic reviews and meta-analyses involve analyzing the results of multiple studies on a particular topic