Slide 1
Slide 1 text
Univariate analysis
SPSS LABORATORY 4
Slide 2
Slide 2 text
1. Calendar
2. Previous lecture recap
3. Descriptive statistics: Univariate analysis
4. Exercise
INDEX
Slide 3
Slide 3 text
Still having problems installing SPSS?
○ Book an appointment for office hour (Thursday, 12.30 - 1.20)
○ Tech problem? Email [email protected]
No extensions for technical issues.
Photo: Startup Weekend Hackathon. Nov.2014
…BEFORE WE START
Slide 4
Slide 4 text
No content
Slide 5
Slide 5 text
Photo: Startup Weekend Hackathon. Nov.2014
TOPICS COVERED SO FAR
■ Import a dataset in SPSS
■ Set variable parameters (Name, Type, Label, Measure)
■ Recoding variables
Slide 6
Slide 6 text
Photo: Startup Weekend Hackathon. Nov.2014
LAUNCH SPSS
■ Go to Canvas
○ Assignments > SPSS Lab 4 - Descriptive statistics - Univariate analysis
○ Download the Week-4.sav file
■ Open the file on SPSS
○ Launch SPP
■ File > Open > Data…
■ Find and open the Week-4.sav file
Slide 7
Slide 7 text
Photo: Startup Weekend Hackathon. Nov.2014
CENTRALITY AND DISPERSION
8
How to calculate the main central tendency and dispersion measures in SPSS.
Centrality:
1. Mean (average)
2. Median
3. Mode
Dispersion:
1. Standard deviation
2. Min and max
3. Range (Max - Min)
Slide 8
Slide 8 text
Photo: Startup Weekend Hackathon. Nov.2014
CALCULATING CENTRALITY AND DISPERSION IN SPSS
Slide 9
Slide 9 text
Photo: Startup Weekend Hackathon. Nov.2014
CALCULATING CENTRALITY AND DISPERSION IN SPSS
Slide 10
Slide 10 text
Photo: Startup Weekend Hackathon. Nov.2014
∑ [tempo]
1239
number of songs
in the dataset.
Population, or n
sum of all values in
the tempo colum
CALCULATING CENTRALITY AND DISPERSION IN SPSS
Numeric variables
Slide 11
Slide 11 text
Photo: Startup Weekend Hackathon. Nov.2014
the tempo of the
602nd song (ordered
from the slowest to
the fastest)
Why 602?
CENTRALITY AND DISPERSION
601 songs
median (song
n.602)
601 songs
Numeric variables
fastest
slowest
Slide 12
Slide 12 text
Photo: Startup Weekend Hackathon. Nov.2014
The most “popular”
tempo.
CALCULATING CENTRALITY AND DISPERSION IN SPSS
Numeric variables
Slide 13
Slide 13 text
Photo: Startup Weekend Hackathon. Nov.2014
Tempo of the fastest
and of the slowest
songs
CALCULATING CENTRALITY AND DISPERSION IN SPSS
Numeric variables
Slide 14
Slide 14 text
Photo: Startup Weekend Hackathon. Nov.2014
CALCULATING CENTRALITY AND DISPERSION IN SPSS
Max - Min
Numeric variables
Slide 15
Slide 15 text
Photo: Startup Weekend Hackathon. Nov.2014
QUESTION: WHAT % OF SONGS HAVE TEMPO > 147 BPM?
Slide 16
Slide 16 text
Photo: Startup Weekend Hackathon. Nov.2014
122 147
97
72
47 172 197
QUESTION: WHAT % OF SONGS HAVE TEMPO > 147 BPM?
Slide 17
Slide 17 text
Photo: Startup Weekend Hackathon. Nov.2014
122 147
97
72
47 172 197
QUESTION: WHAT % OF SONGS HAVE TEMPO > 147 BPM?
Slide 18
Slide 18 text
Photo: Startup Weekend Hackathon. Nov.2014
WHAT ABOUT STRING (TEXT) VARIABLES?
Try to calculate centrality and dispersion of the variable “Genre”.
Centrality:
1. Mean (average)
2. Median
3. Mode
Dispersion:
1. Standard deviation
2. Min and max
3. Range (Max - Min)
Slide 19
Slide 19 text
Photo: Startup Weekend Hackathon. Nov.2014
PERCENT Vs VALID PERCENT
Percent: % calculated on the entire
dataset population (n). In this case,
1,239
Valid Percent: % calculated on the
entire dataset population (n) minus
missing values. In this case,
1,239 - 34 = 1,205
Slide 20
Slide 20 text
Photo: Startup Weekend Hackathon. Nov.2014
MULTIPLE ANALYSES AT ONCE
Analyzing more than one variable at once it might be useful
to conduct comparative analysis. Especially if the variables
share the same scale and measurement unit.
To analyze more than one variables at once, simply add
more variables to the “Variable” box in the Frequencies
window.
This week’s exercise: Analyze loudness and energy. Take a
screenshot of the Frequency table and upload it on Canvas.
Slide 21
Slide 21 text
THANK YOU
Alberto Lusoli
[email protected]
Office hour: Thursday, 12.30pm - 1.20pm (please book an appointment in advance via email).