Statistic summary – Test 2 – Year 2, semester 2.2
Test material – Verhoeven, Statistics in steps
Chapters: 1, 2.1, 2.5, 3.1-2, 5.1-5, 6.1-5, 7.1-2, 8.1-4, 9.1.
∇ Errors
∇ Make sure you know what the variables mean
In mathematics and research, a variable is the designation for any element of a collection. It
is said that the variable traverses the set, or that the variable takes values in that set. The
values a variable could take can be divided into two categories:
Numerical data VS. Non-numerical data
Quantitative data Numbers, big amounts of data that are not very distinctive. Methods to
measure could be questionnaires, experiments or big data files.
Qualitative data Background reasons, explanations that are deeper into the subject.
Methods to measure could be interviews, observations or desk research.
∇ Make sure you know how the variables are measured
The measurement level of a variable indicates the degree to which you can use the values
that you have assigned to the categories in calculations. This classification takes place in four
measurement levels, ascending from ‘low’ to ‘high’: Nominal, Ordinal, Interval and Ratio.
The measurement level is the type of rating scale, it is a precondition for choosing you
analysis technique.
1
,Nominal Discrete variables
(qualitative). For example gender,
residence, school degrees.
Ordinal Ranked variables(qualitative).
For example, agree or disagree, level of
education.
Interval Numerical values that have
no meaningful 0 (quantitative). For
example, degrees Celsius or the Likert
scale. Intervals or values have a certain
relation to each other that is not Figure 1
meaningful so they are equal values.
Ratio Numerical values that have a meaningful 0 and you can calculate proportions
between these values. For example, hours worked.
SPSS options are nominal, scale and ordinal. Scale is used for both
interval and ratio variables. See figure 2 for SPSS display.
Measures of variability
Range The whole range of observations, from minimum to maximum. Figure 2
For example the oldest person you asked the age of is 65 and the
youngest 18, the range is (65-18 =)47. Definition is the difference between the minimum
and maximum scores in a set.
Interquartile range Is a
measure that you use to look at
the middle 50% of your
observations and can be used
for variables from ordinal
measurement. A boxplot (see
figure 3) is used to visualize
what the middle part is and up
until which point the outliers
lie. Definition is the difference
between the 75th and 25th
percentile in a distribution, in Figure 3
other words the middle 50% of
all observations.
Variance The measure that indicates how observations are distributed compared to the
mean. The average quadratic deviation of the mean.
2
, Normal distribution example with standard deviation
∇ Create graphs of minimally four variables that have your interest to visualize the data
To make a graph, go to
the upper part in SPSS
to: Graph Chart
builder and click OK
The next popup will
show where you can
choose from various
graphs, plots and
charts. Under Gallery
(highlighted) you can
choose more charts.
Drag the variables onto
the desired axes and
when done, click OK.
3
Test material – Verhoeven, Statistics in steps
Chapters: 1, 2.1, 2.5, 3.1-2, 5.1-5, 6.1-5, 7.1-2, 8.1-4, 9.1.
∇ Errors
∇ Make sure you know what the variables mean
In mathematics and research, a variable is the designation for any element of a collection. It
is said that the variable traverses the set, or that the variable takes values in that set. The
values a variable could take can be divided into two categories:
Numerical data VS. Non-numerical data
Quantitative data Numbers, big amounts of data that are not very distinctive. Methods to
measure could be questionnaires, experiments or big data files.
Qualitative data Background reasons, explanations that are deeper into the subject.
Methods to measure could be interviews, observations or desk research.
∇ Make sure you know how the variables are measured
The measurement level of a variable indicates the degree to which you can use the values
that you have assigned to the categories in calculations. This classification takes place in four
measurement levels, ascending from ‘low’ to ‘high’: Nominal, Ordinal, Interval and Ratio.
The measurement level is the type of rating scale, it is a precondition for choosing you
analysis technique.
1
,Nominal Discrete variables
(qualitative). For example gender,
residence, school degrees.
Ordinal Ranked variables(qualitative).
For example, agree or disagree, level of
education.
Interval Numerical values that have
no meaningful 0 (quantitative). For
example, degrees Celsius or the Likert
scale. Intervals or values have a certain
relation to each other that is not Figure 1
meaningful so they are equal values.
Ratio Numerical values that have a meaningful 0 and you can calculate proportions
between these values. For example, hours worked.
SPSS options are nominal, scale and ordinal. Scale is used for both
interval and ratio variables. See figure 2 for SPSS display.
Measures of variability
Range The whole range of observations, from minimum to maximum. Figure 2
For example the oldest person you asked the age of is 65 and the
youngest 18, the range is (65-18 =)47. Definition is the difference between the minimum
and maximum scores in a set.
Interquartile range Is a
measure that you use to look at
the middle 50% of your
observations and can be used
for variables from ordinal
measurement. A boxplot (see
figure 3) is used to visualize
what the middle part is and up
until which point the outliers
lie. Definition is the difference
between the 75th and 25th
percentile in a distribution, in Figure 3
other words the middle 50% of
all observations.
Variance The measure that indicates how observations are distributed compared to the
mean. The average quadratic deviation of the mean.
2
, Normal distribution example with standard deviation
∇ Create graphs of minimally four variables that have your interest to visualize the data
To make a graph, go to
the upper part in SPSS
to: Graph Chart
builder and click OK
The next popup will
show where you can
choose from various
graphs, plots and
charts. Under Gallery
(highlighted) you can
choose more charts.
Drag the variables onto
the desired axes and
when done, click OK.
3
