Quantitative Data Analysis 1
Lecture 1 03/02/20
Why statistics / quantitative data analysis?
Research for science and decision making
Data and software are accessible (e.g. SPSS)
Population and sample
Population
o The complete group of interest
o All values within the whole group of interest
Sample
o A subset of the population for which observations are gathered
o The observed values
Random sampling
o Each member of the population has equal chance to enter the sample
Definition of variable types
The statistical analysis you can do depends on the type of variable
Quantitative (“scale variable”): measure a number (by nature)
o Continuous: interval of possible values
(idealised view because in practice you always have to round the number)
E.g. temperature in degrees Celsius, income in Euros
Differences have meaning; hence also called “interval variable”
o Discrete: series of isolated possible values
E.g. number of cars sold (0, 1, 2, 3, …), change in number of employees (…, -2, -1, 0,
1, 2 …)
o No clear division line between discrete and continuous
Qualitative: measure a category
o Ordinal: ordered categories
E.g. small/medium/large drink, job skill (very low, low, medium, high, very high)
o Nominal: unordered categories
E.g. employed/unemployed, brand of a product
Hierarchy in level of information
o Continuous – Discrete – Ordinal – Nominal
Likert variable: used to measure judged
o E.g. agreement: 1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
o A Likert variable is ordinal, not quantitative
o Still, in research it is often treated as quantitative. (Why?)
This presupposes equal distances between successive categories
This is justified if the categories are consistent with equal distances and a
quantitative scale with numbers is shown in the questionnaire.
Exploring your data
For qualitative data (categorical) we use:
o Frequency table
o Bar chart
To see and compare the frequencies and to see the order of categories if ordinal
o Pie chart
To see the shares
o Mode (most frequent outcome)
o Median, only for ordinal data (middle outcome)
For quantitative data we use:
Lecture 1 03/02/20
Why statistics / quantitative data analysis?
Research for science and decision making
Data and software are accessible (e.g. SPSS)
Population and sample
Population
o The complete group of interest
o All values within the whole group of interest
Sample
o A subset of the population for which observations are gathered
o The observed values
Random sampling
o Each member of the population has equal chance to enter the sample
Definition of variable types
The statistical analysis you can do depends on the type of variable
Quantitative (“scale variable”): measure a number (by nature)
o Continuous: interval of possible values
(idealised view because in practice you always have to round the number)
E.g. temperature in degrees Celsius, income in Euros
Differences have meaning; hence also called “interval variable”
o Discrete: series of isolated possible values
E.g. number of cars sold (0, 1, 2, 3, …), change in number of employees (…, -2, -1, 0,
1, 2 …)
o No clear division line between discrete and continuous
Qualitative: measure a category
o Ordinal: ordered categories
E.g. small/medium/large drink, job skill (very low, low, medium, high, very high)
o Nominal: unordered categories
E.g. employed/unemployed, brand of a product
Hierarchy in level of information
o Continuous – Discrete – Ordinal – Nominal
Likert variable: used to measure judged
o E.g. agreement: 1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
o A Likert variable is ordinal, not quantitative
o Still, in research it is often treated as quantitative. (Why?)
This presupposes equal distances between successive categories
This is justified if the categories are consistent with equal distances and a
quantitative scale with numbers is shown in the questionnaire.
Exploring your data
For qualitative data (categorical) we use:
o Frequency table
o Bar chart
To see and compare the frequencies and to see the order of categories if ordinal
o Pie chart
To see the shares
o Mode (most frequent outcome)
o Median, only for ordinal data (middle outcome)
For quantitative data we use:
