Private Hypothesis Testing for Social Sciences

While running any experiment, we often have to consider the statistical power to ensure an effective study. Statistical power or power ensures that we can observe an effect with high probability if such a true effect exists. However, several studies lack the appropriate planning for determining the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Mulay, Ajinkya K, Lane, Sean, Hennes, Erin
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Mulay, Ajinkya K
Lane, Sean
Hennes, Erin
description While running any experiment, we often have to consider the statistical power to ensure an effective study. Statistical power or power ensures that we can observe an effect with high probability if such a true effect exists. However, several studies lack the appropriate planning for determining the optimal sample size to ensure adequate power. Thus, careful planning ensures that the power remains high even under high measurement errors while keeping the type 1 error constrained. We study the impact of differential privacy on experiments and theoretically analyze the change in sample size required due to the Gaussian mechanisms. Further, we provide an empirical method to improve the accuracy of private statistics with simple bootstrapping.
doi_str_mv 10.48550/arxiv.2205.05522
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2205_05522</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2205_05522</sourcerecordid><originalsourceid>FETCH-LOGICAL-a672-4950bb7da1d2872b7d0bab115aa6603476364d6467074641e6594e694ebfe9ab3</originalsourceid><addsrcrecordid>eNotzsFugzAQBFBfeqjSfkBP9Q9AbWOvwzGK2iYSUivBHa1haS0RiGyEyt8nIT2MZk6jx9iLFKneGiPeMPz5OVVKmFQYo9QjS7-Dn3EifljO4_RL0UdeUZz88MO7MfBybDz2vGw8DQ3FJ_bQYR_p-b83rPp4r_aHpPj6PO53RYJgVaJzI5yzLcpWba26LuHQSWkQAUSmLWSgW9BghdWgJYHJNcE1rqMcXbZhr_fbFVyfgz9hWOobvF7h2QVndzxt</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Private Hypothesis Testing for Social Sciences</title><source>arXiv.org</source><creator>Mulay, Ajinkya K ; Lane, Sean ; Hennes, Erin</creator><creatorcontrib>Mulay, Ajinkya K ; Lane, Sean ; Hennes, Erin</creatorcontrib><description>While running any experiment, we often have to consider the statistical power to ensure an effective study. Statistical power or power ensures that we can observe an effect with high probability if such a true effect exists. However, several studies lack the appropriate planning for determining the optimal sample size to ensure adequate power. Thus, careful planning ensures that the power remains high even under high measurement errors while keeping the type 1 error constrained. We study the impact of differential privacy on experiments and theoretically analyze the change in sample size required due to the Gaussian mechanisms. Further, we provide an empirical method to improve the accuracy of private statistics with simple bootstrapping.</description><identifier>DOI: 10.48550/arxiv.2205.05522</identifier><language>eng</language><subject>Computer Science - Cryptography and Security ; Statistics - Methodology</subject><creationdate>2022-05</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,778,883</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2205.05522$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2205.05522$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Mulay, Ajinkya K</creatorcontrib><creatorcontrib>Lane, Sean</creatorcontrib><creatorcontrib>Hennes, Erin</creatorcontrib><title>Private Hypothesis Testing for Social Sciences</title><description>While running any experiment, we often have to consider the statistical power to ensure an effective study. Statistical power or power ensures that we can observe an effect with high probability if such a true effect exists. However, several studies lack the appropriate planning for determining the optimal sample size to ensure adequate power. Thus, careful planning ensures that the power remains high even under high measurement errors while keeping the type 1 error constrained. We study the impact of differential privacy on experiments and theoretically analyze the change in sample size required due to the Gaussian mechanisms. Further, we provide an empirical method to improve the accuracy of private statistics with simple bootstrapping.</description><subject>Computer Science - Cryptography and Security</subject><subject>Statistics - Methodology</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzsFugzAQBFBfeqjSfkBP9Q9AbWOvwzGK2iYSUivBHa1haS0RiGyEyt8nIT2MZk6jx9iLFKneGiPeMPz5OVVKmFQYo9QjS7-Dn3EifljO4_RL0UdeUZz88MO7MfBybDz2vGw8DQ3FJ_bQYR_p-b83rPp4r_aHpPj6PO53RYJgVaJzI5yzLcpWba26LuHQSWkQAUSmLWSgW9BghdWgJYHJNcE1rqMcXbZhr_fbFVyfgz9hWOobvF7h2QVndzxt</recordid><startdate>20220507</startdate><enddate>20220507</enddate><creator>Mulay, Ajinkya K</creator><creator>Lane, Sean</creator><creator>Hennes, Erin</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20220507</creationdate><title>Private Hypothesis Testing for Social Sciences</title><author>Mulay, Ajinkya K ; Lane, Sean ; Hennes, Erin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a672-4950bb7da1d2872b7d0bab115aa6603476364d6467074641e6594e694ebfe9ab3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Cryptography and Security</topic><topic>Statistics - Methodology</topic><toplevel>online_resources</toplevel><creatorcontrib>Mulay, Ajinkya K</creatorcontrib><creatorcontrib>Lane, Sean</creatorcontrib><creatorcontrib>Hennes, Erin</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Mulay, Ajinkya K</au><au>Lane, Sean</au><au>Hennes, Erin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Private Hypothesis Testing for Social Sciences</atitle><date>2022-05-07</date><risdate>2022</risdate><abstract>While running any experiment, we often have to consider the statistical power to ensure an effective study. Statistical power or power ensures that we can observe an effect with high probability if such a true effect exists. However, several studies lack the appropriate planning for determining the optimal sample size to ensure adequate power. Thus, careful planning ensures that the power remains high even under high measurement errors while keeping the type 1 error constrained. We study the impact of differential privacy on experiments and theoretically analyze the change in sample size required due to the Gaussian mechanisms. Further, we provide an empirical method to improve the accuracy of private statistics with simple bootstrapping.</abstract><doi>10.48550/arxiv.2205.05522</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2205.05522
ispartof
issn
language eng
recordid cdi_arxiv_primary_2205_05522
source arXiv.org
subjects Computer Science - Cryptography and Security
Statistics - Methodology
title Private Hypothesis Testing for Social Sciences
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-17T01%3A21%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Private%20Hypothesis%20Testing%20for%20Social%20Sciences&rft.au=Mulay,%20Ajinkya%20K&rft.date=2022-05-07&rft_id=info:doi/10.48550/arxiv.2205.05522&rft_dat=%3Carxiv_GOX%3E2205_05522%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true