Robust Statistical Methods for Empirical Software Engineering
Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper ai...
Ausführliche Beschreibung
Autor*in: |
Kitchenham, Barbara [verfasserIn] |
---|
Format: |
Artikel |
---|---|
Sprache: |
Englisch |
Erschienen: |
2016 |
---|
Schlagwörter: |
---|
Anmerkung: |
© The Author(s) 2016 |
---|
Übergeordnetes Werk: |
Enthalten in: Empirical software engineering - Springer US, 1996, 22(2016), 2 vom: 16. Juni, Seite 579-630 |
---|---|
Übergeordnetes Werk: |
volume:22 ; year:2016 ; number:2 ; day:16 ; month:06 ; pages:579-630 |
Links: |
---|
DOI / URN: |
10.1007/s10664-016-9437-5 |
---|
Katalog-ID: |
OLC2071663667 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | OLC2071663667 | ||
003 | DE-627 | ||
005 | 20230503052529.0 | ||
007 | tu | ||
008 | 200819s2016 xx ||||| 00| ||eng c | ||
024 | 7 | |a 10.1007/s10664-016-9437-5 |2 doi | |
035 | |a (DE-627)OLC2071663667 | ||
035 | |a (DE-He213)s10664-016-9437-5-p | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
082 | 0 | 4 | |a 004 |q VZ |
100 | 1 | |a Kitchenham, Barbara |e verfasserin |4 aut | |
245 | 1 | 0 | |a Robust Statistical Methods for Empirical Software Engineering |
264 | 1 | |c 2016 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ohne Hilfsmittel zu benutzen |b n |2 rdamedia | ||
338 | |a Band |b nc |2 rdacarrier | ||
500 | |a © The Author(s) 2016 | ||
520 | |a Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. | ||
650 | 4 | |a Empirical software engineering | |
650 | 4 | |a Statistical methods | |
650 | 4 | |a Robust methods | |
650 | 4 | |a Robust statistical methods | |
700 | 1 | |a Madeyski, Lech |4 aut | |
700 | 1 | |a Budgen, David |4 aut | |
700 | 1 | |a Keung, Jacky |4 aut | |
700 | 1 | |a Brereton, Pearl |4 aut | |
700 | 1 | |a Charters, Stuart |4 aut | |
700 | 1 | |a Gibbs, Shirley |4 aut | |
700 | 1 | |a Pohthong, Amnart |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Empirical software engineering |d Springer US, 1996 |g 22(2016), 2 vom: 16. Juni, Seite 579-630 |w (DE-627)235946516 |w (DE-600)1401304-6 |w (DE-576)102432406 |x 1382-3256 |7 nnns |
773 | 1 | 8 | |g volume:22 |g year:2016 |g number:2 |g day:16 |g month:06 |g pages:579-630 |
856 | 4 | 1 | |u https://doi.org/10.1007/s10664-016-9437-5 |z lizenzpflichtig |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a SYSFLAG_A | ||
912 | |a GBV_OLC | ||
912 | |a SSG-OLC-MAT | ||
912 | |a GBV_ILN_70 | ||
951 | |a AR | ||
952 | |d 22 |j 2016 |e 2 |b 16 |c 06 |h 579-630 |
author_variant |
b k bk l m lm d b db j k jk p b pb s c sc s g sg a p ap |
---|---|
matchkey_str |
article:13823256:2016----::outttsiamtosoeprclo |
hierarchy_sort_str |
2016 |
publishDate |
2016 |
allfields |
10.1007/s10664-016-9437-5 doi (DE-627)OLC2071663667 (DE-He213)s10664-016-9437-5-p DE-627 ger DE-627 rakwb eng 004 VZ Kitchenham, Barbara verfasserin aut Robust Statistical Methods for Empirical Software Engineering 2016 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s) 2016 Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. Empirical software engineering Statistical methods Robust methods Robust statistical methods Madeyski, Lech aut Budgen, David aut Keung, Jacky aut Brereton, Pearl aut Charters, Stuart aut Gibbs, Shirley aut Pohthong, Amnart aut Enthalten in Empirical software engineering Springer US, 1996 22(2016), 2 vom: 16. Juni, Seite 579-630 (DE-627)235946516 (DE-600)1401304-6 (DE-576)102432406 1382-3256 nnns volume:22 year:2016 number:2 day:16 month:06 pages:579-630 https://doi.org/10.1007/s10664-016-9437-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT GBV_ILN_70 AR 22 2016 2 16 06 579-630 |
spelling |
10.1007/s10664-016-9437-5 doi (DE-627)OLC2071663667 (DE-He213)s10664-016-9437-5-p DE-627 ger DE-627 rakwb eng 004 VZ Kitchenham, Barbara verfasserin aut Robust Statistical Methods for Empirical Software Engineering 2016 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s) 2016 Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. Empirical software engineering Statistical methods Robust methods Robust statistical methods Madeyski, Lech aut Budgen, David aut Keung, Jacky aut Brereton, Pearl aut Charters, Stuart aut Gibbs, Shirley aut Pohthong, Amnart aut Enthalten in Empirical software engineering Springer US, 1996 22(2016), 2 vom: 16. Juni, Seite 579-630 (DE-627)235946516 (DE-600)1401304-6 (DE-576)102432406 1382-3256 nnns volume:22 year:2016 number:2 day:16 month:06 pages:579-630 https://doi.org/10.1007/s10664-016-9437-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT GBV_ILN_70 AR 22 2016 2 16 06 579-630 |
allfields_unstemmed |
10.1007/s10664-016-9437-5 doi (DE-627)OLC2071663667 (DE-He213)s10664-016-9437-5-p DE-627 ger DE-627 rakwb eng 004 VZ Kitchenham, Barbara verfasserin aut Robust Statistical Methods for Empirical Software Engineering 2016 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s) 2016 Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. Empirical software engineering Statistical methods Robust methods Robust statistical methods Madeyski, Lech aut Budgen, David aut Keung, Jacky aut Brereton, Pearl aut Charters, Stuart aut Gibbs, Shirley aut Pohthong, Amnart aut Enthalten in Empirical software engineering Springer US, 1996 22(2016), 2 vom: 16. Juni, Seite 579-630 (DE-627)235946516 (DE-600)1401304-6 (DE-576)102432406 1382-3256 nnns volume:22 year:2016 number:2 day:16 month:06 pages:579-630 https://doi.org/10.1007/s10664-016-9437-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT GBV_ILN_70 AR 22 2016 2 16 06 579-630 |
allfieldsGer |
10.1007/s10664-016-9437-5 doi (DE-627)OLC2071663667 (DE-He213)s10664-016-9437-5-p DE-627 ger DE-627 rakwb eng 004 VZ Kitchenham, Barbara verfasserin aut Robust Statistical Methods for Empirical Software Engineering 2016 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s) 2016 Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. Empirical software engineering Statistical methods Robust methods Robust statistical methods Madeyski, Lech aut Budgen, David aut Keung, Jacky aut Brereton, Pearl aut Charters, Stuart aut Gibbs, Shirley aut Pohthong, Amnart aut Enthalten in Empirical software engineering Springer US, 1996 22(2016), 2 vom: 16. Juni, Seite 579-630 (DE-627)235946516 (DE-600)1401304-6 (DE-576)102432406 1382-3256 nnns volume:22 year:2016 number:2 day:16 month:06 pages:579-630 https://doi.org/10.1007/s10664-016-9437-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT GBV_ILN_70 AR 22 2016 2 16 06 579-630 |
allfieldsSound |
10.1007/s10664-016-9437-5 doi (DE-627)OLC2071663667 (DE-He213)s10664-016-9437-5-p DE-627 ger DE-627 rakwb eng 004 VZ Kitchenham, Barbara verfasserin aut Robust Statistical Methods for Empirical Software Engineering 2016 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s) 2016 Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. Empirical software engineering Statistical methods Robust methods Robust statistical methods Madeyski, Lech aut Budgen, David aut Keung, Jacky aut Brereton, Pearl aut Charters, Stuart aut Gibbs, Shirley aut Pohthong, Amnart aut Enthalten in Empirical software engineering Springer US, 1996 22(2016), 2 vom: 16. Juni, Seite 579-630 (DE-627)235946516 (DE-600)1401304-6 (DE-576)102432406 1382-3256 nnns volume:22 year:2016 number:2 day:16 month:06 pages:579-630 https://doi.org/10.1007/s10664-016-9437-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT GBV_ILN_70 AR 22 2016 2 16 06 579-630 |
language |
English |
source |
Enthalten in Empirical software engineering 22(2016), 2 vom: 16. Juni, Seite 579-630 volume:22 year:2016 number:2 day:16 month:06 pages:579-630 |
sourceStr |
Enthalten in Empirical software engineering 22(2016), 2 vom: 16. Juni, Seite 579-630 volume:22 year:2016 number:2 day:16 month:06 pages:579-630 |
format_phy_str_mv |
Article |
institution |
findex.gbv.de |
topic_facet |
Empirical software engineering Statistical methods Robust methods Robust statistical methods |
dewey-raw |
004 |
isfreeaccess_bool |
false |
container_title |
Empirical software engineering |
authorswithroles_txt_mv |
Kitchenham, Barbara @@aut@@ Madeyski, Lech @@aut@@ Budgen, David @@aut@@ Keung, Jacky @@aut@@ Brereton, Pearl @@aut@@ Charters, Stuart @@aut@@ Gibbs, Shirley @@aut@@ Pohthong, Amnart @@aut@@ |
publishDateDaySort_date |
2016-06-16T00:00:00Z |
hierarchy_top_id |
235946516 |
dewey-sort |
14 |
id |
OLC2071663667 |
language_de |
englisch |
fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">OLC2071663667</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230503052529.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">200819s2016 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s10664-016-9437-5</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC2071663667</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-He213)s10664-016-9437-5-p</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Kitchenham, Barbara</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Robust Statistical Methods for Empirical Software Engineering</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2016</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">© The Author(s) 2016</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Empirical software engineering</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Statistical methods</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Robust methods</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Robust statistical methods</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Madeyski, Lech</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Budgen, David</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Keung, Jacky</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Brereton, Pearl</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Charters, Stuart</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gibbs, Shirley</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Pohthong, Amnart</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Empirical software engineering</subfield><subfield code="d">Springer US, 1996</subfield><subfield code="g">22(2016), 2 vom: 16. Juni, Seite 579-630</subfield><subfield code="w">(DE-627)235946516</subfield><subfield code="w">(DE-600)1401304-6</subfield><subfield code="w">(DE-576)102432406</subfield><subfield code="x">1382-3256</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:22</subfield><subfield code="g">year:2016</subfield><subfield code="g">number:2</subfield><subfield code="g">day:16</subfield><subfield code="g">month:06</subfield><subfield code="g">pages:579-630</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">https://doi.org/10.1007/s10664-016-9437-5</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">22</subfield><subfield code="j">2016</subfield><subfield code="e">2</subfield><subfield code="b">16</subfield><subfield code="c">06</subfield><subfield code="h">579-630</subfield></datafield></record></collection>
|
author |
Kitchenham, Barbara |
spellingShingle |
Kitchenham, Barbara ddc 004 misc Empirical software engineering misc Statistical methods misc Robust methods misc Robust statistical methods Robust Statistical Methods for Empirical Software Engineering |
authorStr |
Kitchenham, Barbara |
ppnlink_with_tag_str_mv |
@@773@@(DE-627)235946516 |
format |
Article |
dewey-ones |
004 - Data processing & computer science |
delete_txt_mv |
keep |
author_role |
aut aut aut aut aut aut aut aut |
collection |
OLC |
remote_str |
false |
illustrated |
Not Illustrated |
issn |
1382-3256 |
topic_title |
004 VZ Robust Statistical Methods for Empirical Software Engineering Empirical software engineering Statistical methods Robust methods Robust statistical methods |
topic |
ddc 004 misc Empirical software engineering misc Statistical methods misc Robust methods misc Robust statistical methods |
topic_unstemmed |
ddc 004 misc Empirical software engineering misc Statistical methods misc Robust methods misc Robust statistical methods |
topic_browse |
ddc 004 misc Empirical software engineering misc Statistical methods misc Robust methods misc Robust statistical methods |
format_facet |
Aufsätze Gedruckte Aufsätze |
format_main_str_mv |
Text Zeitschrift/Artikel |
carriertype_str_mv |
nc |
hierarchy_parent_title |
Empirical software engineering |
hierarchy_parent_id |
235946516 |
dewey-tens |
000 - Computer science, knowledge & systems |
hierarchy_top_title |
Empirical software engineering |
isfreeaccess_txt |
false |
familylinks_str_mv |
(DE-627)235946516 (DE-600)1401304-6 (DE-576)102432406 |
title |
Robust Statistical Methods for Empirical Software Engineering |
ctrlnum |
(DE-627)OLC2071663667 (DE-He213)s10664-016-9437-5-p |
title_full |
Robust Statistical Methods for Empirical Software Engineering |
author_sort |
Kitchenham, Barbara |
journal |
Empirical software engineering |
journalStr |
Empirical software engineering |
lang_code |
eng |
isOA_bool |
false |
dewey-hundreds |
000 - Computer science, information & general works |
recordtype |
marc |
publishDateSort |
2016 |
contenttype_str_mv |
txt |
container_start_page |
579 |
author_browse |
Kitchenham, Barbara Madeyski, Lech Budgen, David Keung, Jacky Brereton, Pearl Charters, Stuart Gibbs, Shirley Pohthong, Amnart |
container_volume |
22 |
class |
004 VZ |
format_se |
Aufsätze |
author-letter |
Kitchenham, Barbara |
doi_str_mv |
10.1007/s10664-016-9437-5 |
dewey-full |
004 |
title_sort |
robust statistical methods for empirical software engineering |
title_auth |
Robust Statistical Methods for Empirical Software Engineering |
abstract |
Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. © The Author(s) 2016 |
abstractGer |
Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. © The Author(s) 2016 |
abstract_unstemmed |
Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method. © The Author(s) 2016 |
collection_details |
GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT GBV_ILN_70 |
container_issue |
2 |
title_short |
Robust Statistical Methods for Empirical Software Engineering |
url |
https://doi.org/10.1007/s10664-016-9437-5 |
remote_bool |
false |
author2 |
Madeyski, Lech Budgen, David Keung, Jacky Brereton, Pearl Charters, Stuart Gibbs, Shirley Pohthong, Amnart |
author2Str |
Madeyski, Lech Budgen, David Keung, Jacky Brereton, Pearl Charters, Stuart Gibbs, Shirley Pohthong, Amnart |
ppnlink |
235946516 |
mediatype_str_mv |
n |
isOA_txt |
false |
hochschulschrift_bool |
false |
doi_str |
10.1007/s10664-016-9437-5 |
up_date |
2024-07-04T03:57:04.417Z |
_version_ |
1803619322480820224 |
fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">OLC2071663667</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230503052529.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">200819s2016 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s10664-016-9437-5</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC2071663667</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-He213)s10664-016-9437-5-p</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Kitchenham, Barbara</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Robust Statistical Methods for Empirical Software Engineering</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2016</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">© The Author(s) 2016</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or more samples. When the distribution of the data differs among groups, or we have ordinal scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based ANOVA-like method.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Empirical software engineering</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Statistical methods</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Robust methods</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Robust statistical methods</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Madeyski, Lech</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Budgen, David</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Keung, Jacky</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Brereton, Pearl</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Charters, Stuart</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gibbs, Shirley</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Pohthong, Amnart</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Empirical software engineering</subfield><subfield code="d">Springer US, 1996</subfield><subfield code="g">22(2016), 2 vom: 16. Juni, Seite 579-630</subfield><subfield code="w">(DE-627)235946516</subfield><subfield code="w">(DE-600)1401304-6</subfield><subfield code="w">(DE-576)102432406</subfield><subfield code="x">1382-3256</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:22</subfield><subfield code="g">year:2016</subfield><subfield code="g">number:2</subfield><subfield code="g">day:16</subfield><subfield code="g">month:06</subfield><subfield code="g">pages:579-630</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">https://doi.org/10.1007/s10664-016-9437-5</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">22</subfield><subfield code="j">2016</subfield><subfield code="e">2</subfield><subfield code="b">16</subfield><subfield code="c">06</subfield><subfield code="h">579-630</subfield></datafield></record></collection>
|
score |
7.399288 |