BAPS Unit 2

Patterns in Time

Does a peak in the explanatory variable occur in concert with a peak in the response variable, with an appropriate time separation

prevalence

is a rare exposure followed by a rare effect

context

is there plausible connection between the two

larger body of research

are there consistent results across many studies

ecological study

observational

“subjects” are groups of people instead of individuals

compares the health of populations in different places or times

may lead to subsequent, better studies on an individual level

case report

observational

researchers write a report about a case or patient in which there was some difficulty in reaching a diagnosis, and that provides a teaching point

new issues and ideas are introduced to the research community and may be the beginnings of a subsequent, better study

cross-sectional study

observational

researchers look at a group of people who fit the general description of the type of subject they’re interested in

a ONE-TIME measurement of exposure and outcome

used to estimate the prevalence of diseases

case-control study

observational

begins with people who have a particular disease (“cases”) then finds a similar group fo healthy people (“controls”)

researchers then look back in time (ask subjects to remember habits) to see which subjects in each group had exposure to the factor of interest; compare the frequency of the exposure in the two groups

retrospective, quick and efficient for studying rare diseases

cohort study

observational

begins with people who have been exposed to a factor of interest

subjects are followed and evaluated for whether they develop the condition of interest

prospective, quick and efficient for studying rare exposures

randomized control trial

experiment

subjects are randomly assigned to a treatment to see the effects on a response variable

other variables are controlled or blocked

can demonstrate cause-and-effectsy

systematic review

identify all relevant studies and synthesizes all the conclusions

meta-analysis

combines the data from multiple similarly designed studies and performs a single statistical analysis

increases the power of the test, but it can be difficult to find studies with similar designs

control, block, randomize

____ what you can

_____ on what you can’t control

_________ the rest

independent and identically distributed

iid stands for

normal distribution

Y - N(mean, st. dev.)

as the response variable is a random variable that has a _____ _____ with mean mu and st. dev. sigma

noise

blocking reduces the “_____” in the response variable

random variability in the measurement of

what does E (epsilon) represent?

sentence fragment

experimental unit

the smallest part of experimental material that is assigned to a factor-level combination within a study (could be a time slot)

nuisance factor

a factor that has an effect on the response but is of no interest to the experimenter; however, the variability it creates in the response needs to be minimized or explained; we block on it

crossed factors / two-way factorial design

two factors are crossed if all possible combinations of levels are randomly assigned to groups of subjects

balanced

the design is _____ if there are the same number of observations for each treatment/cell

repeated measures block

each subject is a block of time slots; the experimental unit is a time slot

usually matched pair

longitudinal experiments

each subject receives one treatment; the same variable is measured on the same subjects at different points in time

crossover experiments

each subject receives more than 1 treatment; subjects are randomly assigned to one of the several orders for receiving the treatment levels

split-plot experiment

randomly assign subjects to one of several levels of the first variable; then, randomly determine the order of presentation of levels of the 2nd variable; experimental units receive all levels of the 2nd variable, but only one level of the 1st variable