Unit 3

Question 1

what is a population?

Answer 1

a collection of objects about which we want information

Question 2

what is a sample?

Answer 2

a part of the population consisting of units we actually examine to gather information

Question 3

what are the two types of sampling that is bias?

Answer 3

voluntary response samples and convenience samples

Question 4

what is voluntary response sampling?

Answer 4

when people make the choice of whether or not they want to participate

Question 5

what is the problem with voluntary response sampling? how is it bias?

Answer 5

usually only people with very strong opinions will bother to phone in and vote, others that dont care will exclude themselves from the sample

Question 6

what is an example of voluntary response sampling?

Answer 6

newspaper publishes a survey, there is a phone number to phone in and vote

Question 7

what is convenience sampling?

Answer 7

when you select a sample based on how convenient it is for you (the one gathering the sample)

Question 8

what is the problem with convenience sampling? how is it bias?

Answer 8

not everyone has the chance of being selected

Question 9

what is an example of convenience sampling?

Answer 9

if we wanted to know if the students like the hours of the library and we stand in university centre and asked the students that were walking by

Question 10

how do we overcome bias?

Answer 10

select a sample by “chance”

Question 11

what are the 5 unbiased forms of sampling?

Answer 11

simple random sampling
stratified
multistage
systematic
census

Question 12

what is the strong statement of simple random sampling?

Answer 12

SRS of size n consists of n individuals from the population chosen in such a way that every group of n individuals has an equal chance to be the sample actually selected

Question 13

what is the weak statement of simple random sampling?

Answer 13

using simple random sample guarantees that every unit in the population has an equal chance of being chosen for the sample

Question 14

what is the definition of stratified random sampling?

Answer 14

we divide the population into strata, then we choose an SRS from within each stratum to form our full sample

Question 15

what is an example of stratified random sampling?

Answer 15

if we want to know how Canadian’s feel about taxes, there is a natural division to this population between provinces and territories and they all pay a different amount of taxes. we would then take a simple random sample from each stratum (province) and then combine responses

Question 16

what is the benefit of stratified random sampling over an SRS?

Answer 16

SRS can sometimes give you insufficient sample size in some of the strata which affects the ability to make conclusions about those strata

Question 17

when do we use multistage sampling?

Answer 17

if we need to select smaller and smaller groups from the population and we can employ different sampling techniques at each stage

Question 18

what is an example of multistage sampling?

Answer 18

if we want to conduct a poll looking into poverty in a third world country, we first select an SRS of villages from the list of all villages, then we select a sample of blocks from these selected villages, then select a sample of households from these blocks

Question 19

is multistage, just as good as SRS? why?

Answer 19

no, because not all units have the same chance of being included and not every combination of units has a chance of being selected

Question 20

look at the difference between SRS, stratified and multistage *SRS, Stratified and Multistage

Question 21

what is systematic random sample? explain how it works.

Answer 21

in this sampling procedure, we start with a numbered list of all N individuals in the population. to select a sample of n individuals, we randomly select a number from 1 to k, where k= N/n. the sample then consists of the k th individual on the list and every k th individual after that.

Question 22

what is an example of systematic random sample?

Answer 22

a teacher wants to select a random sample of five of her 40 students in her class to participate in a demonstration.
to do this she randomly selects a number from 1 to 8, she selects the student in that position on the class list and every 8 th student after that
ex. she selects the number 3 (randomly) then she will select student 11, then 19, then 27, then 35

Question 23

is a systematic random sample a SRS?

Answer 23

no, because not every group is equally likely to be chosen (for example, the first 8 students wouldn’t all be able to be in a group)

Question 24

what is the biggest advantage of a systematic sample?

Answer 24

it is easy and fast to select
dont always need a list of the entire population

Question 25

in what circumstance would we not use a systematic random sample?

Answer 25

when there is a pattern in the population list (for example, the days of the week)

Question 26

what is census sampling?

Answer 26

collection of data from every individual in the entire pooulation

Question 27

what is the best way to sample?

Answer 27

census (we don't use it because its very costly)

Question 28

what are 3 types of bias?

Answer 28

undercoverage non response response bias

Question 29

what is undercoverage?

Answer 29

list from where the sample is drawn is incomplete ex. you cold call landlines, not everyone has a landline anymore

Question 30

what is non response?

Answer 30

people may refuse to answer mail in surveys have a low rate of return

Question 31

what is response bias?

Answer 31

people may lie they may not remember what their answer was question is poorly worded

Question 32

what are 6 things to remember in selecting a "good sample"?

Answer 32

select the sample in an unbiased manner proper interviewer training non-leading questions try to make non- response not an issue include all population units in your possible sample voluntary response and convenience response are not appropriate

Question 33

broadly speaking what are two ways we can obtain data?

Answer 33

we can sample units from a larger population and just simply observe some characteristic of these sampled units (cant impose a cause and effect relationship) we can carry out an experiment where we observe the response to some treatment imposed on the units (can impose a cause and effect relationship)

Question 34

what is a cause and effect relationship?

Answer 34

Causality is a relationship between two variables, in which one event or process causes an effect on the other event or process. For example, research tells us that there is a positive correlation between ice cream sales and sunburns. Meaning, as ice cream sales increase, so do instances of sunburns.

Question 35

what is the one reason why we wouldn't be able to conclude a cause and effect relationship between two variables?

Answer 35

if its a poorly designed experiment

Question 36

what happens in an observational study?

Answer 36

the researcher simply observes the units or subjects, and measures the variables or characteristics of interest (the researcher DOES NOT actively interfere with the units)

Question 37

what happens in an experimental study?

Answer 37

the researcher imposes some treatment on the units or individuals (imposes the values of x)

Question 38

experimental units

Answer 38

the objects on which our experiment is being conducted. (if the experiment is done on humans, they are referred to as subjects)

Question 39

factors

Answer 39

the explanatory variables (they will explain) in an experiment (ex. amt of sleep you get)

Question 40

levels or factor levels

Answer 40

individual values of the factors

Question 41

treatment

Answer 41

a specific experimental condition applied to the units (they're combinations of factor levels)

Question 42

response variable

Answer 42

the variable that is measured following the application of a treatment

Question 43

what does randomization do to treatment groups?

Answer 43

makes them as similar as possible (this is a good thing)

Question 44

what is a completely randomized design?

Answer 44

an experimental design that randomly assigns treatments to the experimental units

Question 45

lurking variable

Answer 45

anything you're not studying

Question 46

what is an example of a lurking variable?

Answer 46

picking people based on what time they show up to work, the lurking variable would be that the people that show up earlier are normally more motivated

Question 47

why is random allocation important?

Answer 47

because it eliminates the effect of any other lurking variable

Question 48

a well designed experiment implements a control, what is it?

Answer 48

a mechanism to minimize the effects of other variables (besides the factors) on the response variable

Question 49

what are the 4 methods of control?

Answer 49

comparison of two or more treatments in the same environment use of a control group double- blinding blocking

Question 50

when is a treatment group required and not required?

Answer 50

required when there is only one other treatment group not required if there are at least two other treatment groups

Question 51

biased form of sampling example in respect to a new medication

Answer 51

when a medical company wants to show that its new medication really works, they give it to the sickest patients who will tend to show the most improvement

Question 52

double blind experiment

Answer 52

an experiment in which neither the subject nor the person administering the treatment knows which treatment is the one being applied at any time

Question 53

single blind

Answer 53

when only one person knows

Question 54

what does the double blind method reduce?

Answer 54

the placebo effect and biased interpretations of the results

Question 55

replication

Answer 55

repeat each treatment on many subjects so as to reduce variation in the results

Question 56

randomization

Answer 56

using chance to assign experimental units to treatments

Question 57

what is blocking?

Answer 57

a group of experimental units that are similar in a way that may affect the responses

Question 58

randomized block design

Answer 58

when the subjects are divided into blocks and the treatments are assigned at random within the blocks

Question 59

how are completely randomized design and randomized block design different?

Answer 59

randomized block design: randomization occurs within blocks, the blocks are not randomly formed completely randomized design: randomization occurs when assigning the treatment to the experimental units

Question 60

why is a blocking variable not a factor?

Answer 60

because factors are explanatory variables that we would like to study for their impact on the response variable

Question 61

what is the purpose of an experiment?

Answer 61

to avoid confounding and to establish causation

Question 62

a good experiment will.... (7)

Answer 62

-allocate units to treatments randomly -control the effect of possible lurking variables (by comparing different treatments or using control groups) -use replication to apply each treatment to several individuals -draw a diagram of the design to better explain it -eliminate bias by conducting a double-blind experiment if possible -remember the placebo effect when inferring causation -use a randomized block design if some units are similar with respect to the response variable