Unit 4 (3) Vocab - Collecting Data

Question 1

What is stratified random sampling?

Answer 1

In a study, the target population gets split into groups (strata, stratum [singular]) containing subjects with shared characteristics relevant to the goals of the study. Within every strata, subjects are randomly selected for data collection.

This method gives more precise estimates of population values than SRS does.

Question 2

What is simple random sampling (SRS)?

Answer 2

When an SRS is taken from a population, subjects within the population are randomly selected. Everyone in the population has an equal chance of being selected. There must not be replacement after each sampling.

Question 3

What is cluster sampling?

Answer 3

In a study, the researcher randomly selects subsets (clusters) within the population and then takes a census of every cluster. The clusters are usually within close proximity.

Cluster sampling saves time and money since data is collected from samples that are close together.

Question 4

What is an observational study?

Answer 4

An observational study is a study in which individuals are observed over time for data collection without trying to affect them like in an experiment.

Question 5

What is the difference between a retrospective observational study and a prospective observational study?

Answer 5

Retrospective observational studies use or reassess data that already exists, while prospective observational studies collect new data from individuals.

Question 6

What is an experiment?

Answer 6

An experiment is a procedure in which experimental subjects are assigned to treatments to test a hypothesis and analyze variables.

Question 7

What is convenience sampling?

Answer 7

A convenience sample consists of individuals who are easy to reach. In most cases, people in these samples are not representative of the whole population.

Question 8

What is a voluntary response sample?

Answer 8

A voluntary response sample consists of individuals who volunteer to join in response to an open invitation. In most cases, people in these samples are not representative of the whole population.

Question 9

Define undercoverage

Answer 9

Undercoverage occurs when some members of the population are less likely to be chosen or can’t be chosen for a sample. This method can show bias if individuals less likely to be included differ in relevant ways from the other members of the population.

Question 10

Define nonresponse

Answer 10

Nonresponse occurs when an individual chosen for a sample can’t be contacted or refuses to participate. This method can show bias if individuals who don’t respond differ in relevant ways from the other members of the population.

Question 11

Define response bias

Answer 11

Response bias occurs when there is a consistent pattern of inaccurate responses to a survey question. This can be caused by the wording of a question, characteristics of the interviewer, lack of anonymity, or other factors.

Question 12

What is bias? What does it cause?

Answer 12

Bias occurs when a study has been conducted to either underestimate or overestimate data values that you want to know. This can create false statistics and misinformation.

Question 13

What is an explanatory variable? What is a response variable?

Answer 13

Explanatory variables can predict or explain changes in the response variable (independent/x variable), and response variables measure the outcome of a study (dependable/y variable).

Question 14

What is confounding?

Answer 14

Two variables are confounded when their effects on a response variable can’t be distinguished from each other. Due to this, sometimes cause and effect can’t be established.

Question 15

Define treatment

Answer 15

A treatment is a combination of levels and factors (explanatory variables) imposed onto experimental units in an experiment.

Question 16

Define experimental unit/subject

Answer 16

Experimental units (or subjects if they’re human) are treated in experiments.

Question 17

Define placebo

Answer 17

A placebo is a treatment that has no active ingredient but is likewise like other treatments.

Question 18

What is a control group?

Answer 18

A control group in an experiment provides a baseline for comparing the effects of other treatments. It may receive a placebo, active treatment, or no treatment.

Question 19

What is the placebo effect?

Answer 19

The placebo effect describes when a placebo in an experiment causes subjects to believe that they are responding favorably to an inactive treatment.

Question 20

What is a single-blind experiment? What is a double-blind experiment?

Answer 20

In a single-blind experiment, either the subject or the administrator doesn’t know the treatment the subject’s receiving, and in a double-blind experiment, both the subject and the administrator don’t know the treatment the subject’s receiving.

Question 21

What is random assignment, and what does it do?

Answer 21

Random assignment is a chance process to assign experimental units to treatments, or vice versa. This creates roughly equivalent groups before treatments are imposed.

Question 22

What is completely randomized design?

Answer 22

Completely randomized design is when the experimental units are assigned to the treatments completely at random.

Question 23

How do you correctly take a random sample of a population?

Answer 23

Three steps:
1. Number each unit of the population 1-n (n = population size)
2. Randomly generate k numbers needed from n amount of units
3. Gather data from the k number of random units selected from the random sampling

Question 24

What is a block, and what is a randomized block design?

Answer 24

Blocks are groups of experimental units that are similar with respect to the variable that should affect the response. Treatments are assigned at random within each block, and the results of one block are compared with the others to determine the effectiveness of the treatments.

Question 25

Compare blocking and stratified random sampling

Question 26

What is a matched pairs design in an experiment?

Question 27

Compare matched pairs design and completely randomized design

Question 28

What makes the results of an experiment “statistically significant?”

Question 29

What is systematic random sampling?

Answer 29

To conduct a systematic random sample, pick “k” number based on population size and desired sample size, randomly select a value 1-k, and select every kth individual thereafter for the sample.

This sampling method may not always be representative of the whole population, but it is easy to implement.

Question 30

What is replication, and what does it do?

Answer 30

Replication is when each treatment is used on enough experimental units so

Question 31

What is the scope of inference?

Question 32

What is the difference between an observational study and an experiment?

Answer 32

Observational designs survey or gather data from subjects without trying to affect them, while experiments assign subjects to treatments to determine cause and effect.