Lecture 17

Question 1

why meta-analysis is at the top of the pyramid?

Answer 1

from this type of research we get the best evidence and results

Question 2

steps of meta-analysis

Answer 2

• Collect data from publications
• Apply statistical methods
Interpret the results obtained

Question 3

is it combinable, we must answer these questions:

Answer 3

• Do the studies estime a sufficiently similar effect?
• Is the study result sufficiently homogeneous (measurement of the result)?
Do the studies deal with sufficiently comparable/homogeneous populations?

Question 4

if not combinable, the meta-analysis will

Answer 4

• analyse apples and pears
  
  • GIGO (garbage-in, garbage-out) +vb method
  • quality of a study will be omitted
  • Nor should objective conclusions be expected if only published/statistically significant results will be analyzed

Question 5

fixed meta-analysis assumption:

Answer 5

The observed variation in treatment effects in the different studies is due entirely to sampling variation
• The underlying treatment effect is the same in all the study populations.
Sometimes too simple in reality.

Question 6

what happens to the way treatment affects individuals in random-effect model?

Answer 6

the treatment effects of individual studies may differ from each other

Question 7

what is assumed about treatment effect?

Answer 7

the treatment effect is universal, and the meta-analysis provides the best available estimate of it.

Question 8

what is the estimate in random-effect meta-analysis?

Answer 8

the estimate is of a mean effect about which it is assumed that the true study effects vary.

Question 9

what is assumed in fixed-effect meta-analysis?

Answer 9

the true effect is the same in each study and the only reason for variation in the estimates between studies is sampling error. It is assumed that the treatment effect is universal, and the meta-analysis provides the best available estimate of it.

Question 10

what cluster analysis does with data?

Answer 10

Cluster analysis divides data into groups (clusters) that are meaningful, useful, or both.

Question 11

what is cluster analysis?

Answer 11

Cluster analysis groups data objects based only on information found in the data that describes the objects and their relationships.

Question 12

what factors produce better clusters?

Answer 12

the greater the similarity (or homogeneity) within a group and the greater the difference between groups, the better or more distinct the clustering.

Question 13

define hierarchical cluster

Answer 13

a set of nested clusters that are organized as a tree

Question 14

define partitional cluster

Answer 14

a division of the set of data objects into nonoverlapping subsets (clusters) such that each data object is in exactly one subset.

Question 15

what K-means method defines?

Answer 15

defines a prototype in terms of a centroid, which is usually the mean of a group of points, and is typically applied to objects in a continuous n-dimensional space.

Question 16

what K-medoid defines?

Answer 16

defines a prototype in terms of a medoid, which is the most representative point for a group of points, and can be applied to a wide range of data since it requires only a proximity measure for a pair of objects.

Question 17

do centroid and medoid correspond to an actual data point?

Answer 17

centroid does not, but medoid does

Question 18

K-means issues

Answer 18

• Number of clusters is predefined 
	• Initial centroids 
	• Restricted to data for which there is a notion of a centre (centroid) 
	•  Empty clusters 
 Outliers

Question 19

why we use factor analysis?

Answer 19

to reduce the number of variables to explain and to interpret the results

Question 20

how factor analysis reduces data?

Answer 20

It does this by seeking underlying unobservable (latent) variables that are reflected in the observed variables (manifest variables)

Question 21

do we need large sample size for factor analysis?

Answer 21

yes, because this model is based on correlation matrix, which requires large sample size

Question 22

is 50 cases for factor analysis sufficient?

Answer 22

no, to have an excellent analysis, we need to have 1000 + cases

Question 23

explain exploratory (EFA) factor analysis

Answer 23

A method to explore the underlying structure of a set of observed variables

Question 24

explain confirmatory (CFA) factor analysis

Answer 24

A method to verify a factor structure that has already been defined

Question 25

what is the basic assumption of factor analysis

Answer 25

for a collection of observed variables there are a set of underlying variables called factors (smaller than the observed variables), that can explain the interrelationships among those (original) variables.

Question 26

how the total variance is split

Answer 26

common variance and unique variance

Question 27

define common variance

Answer 27

the amount of variance that is shared among a set of items. Items that are highly correlated will share a lot of variance.

Question 28

define unique variance

Answer 28

Unique variance is any portion of variance that’s not common

Question 29

define specific variance

Answer 29

is variance that is specific to a particular item.

Question 30

define error variance

Answer 30

comes from errors of measurement and basically anything unexplained by common or specific variance.

Question 31

what is the main difference between common factor analysis and principle components

Answer 31

The main difference between common factor analysis and principal components is that factor analysis assumes total variance can be partitioned into common and unique variance, whereas principal components assumes common variance takes up all of total variance (i.e., no unique variance).

Question 32

what is the value of total variance for common factor analysis and principle components

Answer 32

total variance is 1