Exam 1 study Flashcards by Brian Horton

what is an observational study

observe only - no interaction or control

How well did you know this?

Not at all

Perfectly

What is an experiment

Exert control to better understand the impact of treatments vs response.

How well did you know this?

Not at all

Perfectly

What are treatments

what we do to the person in our experiment

How well did you know this?

Not at all

Perfectly

What are explanatory variables

a variable that we think explains or causes changes in the response variable

How well did you know this?

Not at all

Perfectly

What are response variables

a variable that measures the outcome or result of a study

How well did you know this?

Not at all

Perfectly

What is the difference between a population and a sample

A population is ALL, a sample is a subset selected for study

How well did you know this?

Not at all

Perfectly

Name the 6 main sampling techniques

Random (SRS - Simple random sample)
Cluster
Stratified
Systematic
Convenience
Voluntary Response Survey

How well did you know this?

Not at all

Perfectly

What is a cluster study

1 group from the population (1 class to represent entire school)

How well did you know this?

Not at all

Perfectly

What is a stratified sample

Proportional sampling of the population (College students could be represented by undergrad and grad students. Sample should have correct percentage of each)

How well did you know this?

Not at all

Perfectly

What is a systematic sample

Every nth person is selected

How well did you know this?

Not at all

Perfectly

What is a convenience sample

Interviewer selects people. Mall survey

How well did you know this?

Not at all

Perfectly

What is a voluntary response survey

Depends on individuals to initiate contact, call in polls, online polls. Not valid

How well did you know this?

Not at all

Perfectly

What is bias?

Repeat the same experiment over and over and get the same WRONG result

How well did you know this?

Not at all

Perfectly

Does sample size effect bias?

NO!

How well did you know this?

Not at all

Perfectly

How do we correct for bias?

change sampling method. Must have demographics for population to reduce bias

How well did you know this?

Not at all

Perfectly

What is sampling error?

cannot be controlled (variability). Repeating the same experiment over and over variations are sampling error.

How can we decrease variability?

Increase sample size

How do you computer MOE?

1/ Square root of sample size

What is a confidence interval

Result + or - MOE

Can sampling error be eliminated?

NO!

What is the difference between a parameter and a statistic?

Parameter refers to population and statistic refers to sample

What is the placebo effect?

Positive response independent of treatment?

What is the purpose of using a placebo?

placebo attempts to reduce bias

What is a lurking variable?

something not being accounted for in the study

What are confounding variables

measuring more than we need. Variables are confounding when their effects on a response variable cannot be distinguished from each other. Confounding variables can be either explanatory or lurking variables

What 3 components are necessary for ethical human studies

Institutional Review Board Informed Consent Confidentiality

What are explanatory variables

Controlling the response (what we are changing)

What are response variables

The outcome of the experiment - what we are measuring

What is reliability

measure same person, same way get same result. Results can be both reliable and biased

What is rate?

change/time - more valid measure than a simple count of occurrences (response/population)

What is a block design (and can you draw one)

random assignment of subjects to treatments is carried out separately within each block. (eg: group (block) patients based on disease and then randomly assign to different treatments)

Does an observational study have a design?

NO!

What is a randomized comparative experiment (and can you draw one)

All subjects are allocated at random among all the treatments. eg: all subjects randomly receive either coke, pepsi or RC cola - response variable is BMI - no blocking

What is a matched pair design?

matched pair compares two treatments eg: does coke or pepsi taste better or it could be two subjects each getting one treatment.

Give 3examples of nonsampling errors

nonresponse, response error(inaccurate response), processing errors

what is a sampling frame

list of the subjects

what is statistical significance?

an observed effect of a size that would rarely occur by chance.

what is qualitative

numbers without meaning (index)

what is quantitative

numbers with meaning