Skip to content
Snippets Groups Projects
Commit 9927a782 authored by Lyudmila Vaseva's avatar Lyudmila Vaseva
Browse files

[Literature] Add notes on Exploratory Data Analysis

parent c053672b
No related branches found
No related tags found
No related merge requests found
......@@ -117,6 +117,16 @@
year = {2016}
}
@article{SanFish1994,
author = {Sanderson, Penelope and Fisher, Carolanne},
title = {Exploratory Sequential Data Analysis: Foundations},
year = {1994},
volume = {9},
url = {http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.380.762&rep=rep1&type=pdf},
journal = {Human-Computer Interaction},
pages = {251–317},
}
@article{Stemler2001,
title = {An overview of content analysis},
author = {Stemler, Steve},
......
......@@ -2665,3 +2665,54 @@ the design. In qualitative research, validation takes the form of triangulation.
"Two fatal flaws that destroy the utility of a content analysis are faulty definitions of
categories and non-mutually exclusive and exhaustive categories."
================================================
\cite{SanFish1994}
citing Tukey 1977 on Exploratory Data Analysis (EDA):
"EDA is a simple, visual, but still quantitative approach to data that
allows an investigator to achieve a richer qualitative understanding by
"looking at data to see what it seems to say" (Tukey, 1977, p. v)."
"EDA helps
researchers to generate hypotheses about what might be happening in a
data set-rather than to test hypotheses, which is confirmatory data analy-
sis."
3 principles
1) Continual openness and reexpression.
2) Initial skepticism.
3) Exploration versus confirmation.
1)
"Instead of immediately imposing
a model on the data, which might conceal important details, EDA analysts
try to find patterns in the data and to describe them with simple summary
statistics."
"It may take several iterations before the analyst achieves a
satisfactory summary or "smoothing" of the data. The so-called smooth
part of a data set is the variability that the analyst has accounted for so far,
whereas the "rough" part is the vadability that remains unexplained."
2)
"Because EDA analysts assume that there is no
uniquely correct numerical summary of a data set, they are very skeptical
of initial numerical summaries."
"To help detect patterns and search for data points that do
not fit the smooth part (outliers), EDA analysts rely heavily on visualiza-
tion."
3)
"EDA distinguishes exploratory from
confirmatory data analysis but always maintains a productive interplay
between them that is complementary rather than antagonistic."
generates hypotheses that can later be tested
--------------------------------
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment