Created by Nikolas Bosin
about 5 years ago
|
||
Question | Answer |
What is a suitable statistical software for large data sets? | Stata |
What is a suitable statistical software for open source? | R |
What is a statisctical software that is graphically oriented? | SPSS |
Three important points about the preperation for data analysis are... | - Coding - Editing - Not to overwrite original dataset |
What is "coding"? | Coding describes process through which survey answers, texts, etc. are transformed into processable numbers |
How do we call answers (e.g. in surveys) that are not processable? | Missing values |
What are values to describe the central point of data? | - Mode (even applicable to nominal variables) - Median (expects at least categorical variables) - Mean (only makes sense for metric variables) |
What are values to describe the spread around the central point? | - Variance - Standard deviation - Range |
Formula for the standard error | SE = s/(√n) |
Formula for covariance | Gebe den Text hier ein... |
Formula for Pearson's correlation coefficient | Gebe den Text hier ein... |
How do you call a score that is obviously deviant from the remainder of the data set? | Outlier |
Why do we use data reduction? | To express the same amount of information with less data |
What are two data reduction techniques? | - Factor analysis (groups variables) - Cluster analysis (groups observations) |
What is "factor analysis"? | Factor analysis tries to identify a set of common underlying dimensions, known as factors, in a group of variables (--> Questions about how confident someone is in a personality test should give information about their openness --> here openness is the factor and the questions are the variables) |
There are two different types of factor analysis, which are ... | - Exploratory factor analysis - Confirmatory factor analysis |
The three steps of exploratory factor analysis are... | 1. Examine which variables are correlated by writing them in a correlation matrix 2. Extract factors from the variables 3. Factors are rotated to maximise the relationships between the variables and some of the factors |
How many factors do we obtain? There are three rules of thumb to tell... | - Kaiser criterion (factors with eigenvalues >1) - Percetage of variance criterion - Elbow criterion |
What to do with the results of a factor analysis? | - Study cross loadings (cross loadings = when one variable loads on multiple factors) - Asses reliability of factors through Cronbach's alpha (>=0,7 is good) |
Whats the formula for Cronbach's alpha? |
Gebe den Text hier ein...
Image:
Alpha (binary/octet-stream)
|
What is the goal of the cluster analysis? | To assign observations to homogeneous clusters |
What clustering algorithms exist? | - K-means clustering - Hirarchical clustering |
Want to create your own Flashcards for free with GoConqr? Learn more.