Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Data Analytics > Methods II > Flashcards

Methods II Flashcards

1
Q

What is sampling

A

The process of selecting entities from a larger population, so that by studying the sample we might infer something about the population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is simple random sampling?

A

select a ample of random entities from the population such that they have no pre-study correlation: e.g. based on IBAN, or phone number.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is stratified random sampling?

A

Sampling such that every subgroup (of intrest) is represented and the possibility of biasing is reduced.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is cluster random sampling?

A

Sampling of a population that is known to be a subgroup of a larger population: e.g. sample only people from a certain city or with a certain occupation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

what is convenience random sampling?

A

The sampling of entities of a population based on what is readily or easily (conveniently) available to the researchers. eg. friends, family, coworkers.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What does the sampling distribution (graph) tells us?

A

How much random error we should expect just from the sampling process.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Is this true?: “a random sample is always representative”

A

No, because a random sample can be anything, so also a set of just extreme cases.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

A sample size of 30 is sufficient for a sampling distribution

A

False, n=30 is when sampling distribution comes close the gaussian distribution. If it is sufficient depends on the study.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

“It must be possible for a hypotheses to be false”

A

True, if it is not falsifiable, there is no point of studying it.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is meant with validity?

A

A systematic error in your study. often introduced by poor measurement equipment or ignorance of environmental influences by experimenter. often CAN BE eliminated if identified.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is meant with reliability?

A

A random error (or noise) in the outcome of your studies. Unpredictable fluctuations that neither can be controlled or explained. Often it is possible to reduce the effect of this error.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

“With statistical analysis we study the systematical error”

A

False, the systematical error is determined by the contours of the study itself. Sys.an. has to do with the random error of the data, and finding an explaination.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the description of a main effect of a response factor?

A

When direct changes in the response factor create a destinct difference in the response, without other variables. e.g. all longer conversations lead to more profit compared to shorter ones, indepent of email or phone communication.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the description of an interaction effect of a response factor?

A

When the effect of an response factor is dependent on an other factor. e.g. although longer conversations earn more profit; longer phones > longer emails while short phone < short email. Hence the communication type is a response factor with an interaction effect.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

main effect of channel

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

Interaction effect

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

main effect of channel, interaction effect

18
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

None

19
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

main effect of length

20
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

main effect of length, interaction effect

21
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

main effect of channel, main effect of length, interaction effect

22
Q

Determine what effects are present: (main effect of channel, main effect of length, interaction effect)

A

main effect of channel, main effect of length

23
Q

In factor structures, what is meant with “A crossed with B”?

A

Any level of B can co-occur with any level of A

24
Q

In factor structures, what is meant with “A nestedd with B”?

A

a level of B can co-occur with just one level of A

25
Q

Is the following factor structured nested or crossed? “Every teacher may grade in any course”

A

crossed

26
Q

Is the following factor structured nested or crossed? “Every teacher may grade in just one course”

A

nested

27
Q

What research design allows us to study all main effects and interaction effects?

A

factorial design

28
Q

What is a full factorial design?

A

One run for every possible combination of factors and levels.

29
Q

When do we use fractional factorial design?

A

Either we can’t afford a full design, we’re not interested in all factors, or some combinations are not possible.

30
Q

What is aliasing in research design?

A

When a full factorial design is not possible, it is not possible to estimate all main and interaction effects. So some effects are confounding.

31
Q

What type of fractional design is depicted?

A

Main effects aliased with interactions

32
Q

What type of fractional design is depicted?

A

Main effects aliased with eachother.

33
Q

What is a resolution II design?

A

Main effects confounded with other main effects. Useless, as no interaction can be derived.

34
Q

What is a resolution III design?

A

Main effects confounded with two-factor interactions.

35
Q

What is a resolution IV design?

A

Main effects confounded with two-factor interactions. + Two-factor interactions confounded with other two-factor.

36
Q

What is a resolution V design?

A

Main effects and two-factor interactions unconfounded, etc…

37
Q

In research design, what is the general principle of Random assignment?

A

By randomly assigning patients to treatments, so that effects of other factors effect both treatments equally.

38
Q

In research design, what is the general principle of Replication?

A

Repeat the study to reduce variance and random error.

39
Q

In research design, what is the general principle of blocking?

A

Distribute samples across different subgroups such that the bias from a batch of samples is reduced over all subgroups. Important here is that the bias might be known, but cannot be controlled.

40
Q

In research design, what is the general principle of balance?

A

Asure the same number of samples per sample group. This simplifies the analysis and reduces proportion errors.

Data Analytics (9 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions