{"id":101872,"date":"2016-01-07T12:48:07","date_gmt":"2016-01-07T17:48:07","guid":{"rendered":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/2016\/01\/07\/appendix-c-sensitivity-to-the-turnout-forecast\/"},"modified":"2024-04-14T04:14:40","modified_gmt":"2024-04-14T09:14:40","slug":"appendix-c-sensitivity-to-the-turnout-forecast","status":"publish","type":"post","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-c-sensitivity-to-the-turnout-forecast\/","title":{"rendered":"Appendix C: Sensitivity to the turnout forecast"},"content":{"rendered":"<figure class=\"wp-block-image alignright\"><a href=\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/?attachment_id=276443\"><img decoding=\"async\" class=\"wp-image-276443\" src=\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/2016\/01\/PM_15.01.07_voterModels_sensitivity420px.png\" alt=\"Sensitivity of Perry-Gallup index\" ><\/a><\/figure>\n\n<p class=\"wp-block-paragraph\">The candidate preferences of voters and nonvoters in 2014 were very different. This fact makes cutoff methods very sensitive to the chosen turnout threshold. Using the Perry-Gallup method, the forecast margin ranges from a tie vote (47%-47%) with a more inclusive model (a turnout forecast of 60% of registered voters, 42% of the general public) to a 5 percentage point Republican advantage with a more restricted model (a forecast of 45% turnout among registered voters,\u00a0 33% of the general public).\nUsing the vote among verified voters as a yardstick for where voter preferences stood at that time, a turnout forecast of 50% of registered voters (35% of the general public) comes closest to the benchmark, yielding a Republican lead of 3 points (49% to 46%).<\/p>\n\n<p class=\"wp-block-paragraph\">Given that <a href=\"http:\/\/www.electproject.org\/2014g\" target=\"_blank\" rel=\"noopener noreferrer\">estimates<\/a> for the 2014 general election put the national turnout at 35.9% of the voting eligible population, it would seem that this is the correct cutoff to use. But it is clear that the survey sample used here \u2013 registered voters who completed two waves of interviews and were matched to the voter file \u2013 overrepresented likely voters, since 63% of them (not 50%) are verified as having voted. As a result, a 60% turnout cutoff for registered voters (42% of the general population) was used for analysis.<\/p>\n\n<p class=\"wp-block-paragraph\">Most survey samples are likely to have a similar bias, if not to the same extent. The problem is that it is difficult to know how much a given sample overrepresents likely voters. For this sample in this election, the Perry-Gallup method produces a forecast that is too Democratic when the appropriate cutoff is employed.<\/p>","protected":false},"excerpt":{"rendered":"<p>The candidate preferences of voters and nonvoters in 2014 were very different. This fact makes cutoff methods very sensitive to the chosen turnout threshold. Using the Perry-Gallup method, the forecast margin ranges from a tie vote (47%-47%) with a more inclusive model (a turnout forecast of 60% of registered voters, 42% of the general public) [&hellip;]<\/p>\n","protected":false},"author":294,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"sub_headline":"","sub_title":"","_crdt_document":"","_prc_public_revisions":[],"bylines":[],"acknowledgements":[],"displayBylines":true,"relatedPosts":[],"reportMaterials":[],"multiSectionReport":[],"package_parts__enabled":false,"package_parts":[],"_prc_fork_parent":0,"_prc_fork_status":"","_prc_active_fork":0,"datacite_doi":"","datacite_doi_citation":"","_prc_seo_qr_attachment_id":0,"spoken_article_player_enabled":true,"footnotes":""},"categories":[43,42,36,57,25,357,39,49,50],"tags":[],"bylines":[],"collection":[],"datasets":[],"level_of_effort":[],"primary_audience":[],"information_type":[],"_post_visibility":[],"formats":[458],"_fund_pool":[],"languages":[],"regions-countries":[],"research-teams":[528],"class_list":["post-101872","post","type-post","status-publish","format-standard","hentry","category-election-2014","category-election-2016","category-methodological-research","category-political-civic-engagement","category-politics-policy","category-survey-methods","category-us-elections-voters","category-voter-demographics","category-voter-participation","formats-report","research-teams-methods"],"label":false,"post_parent":101878,"word_count":251,"canonical_url":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-c-sensitivity-to-the-turnout-forecast\/","art_direction":false,"_embeds":[],"table_of_contents":[{"id":101878,"title":"Can Likely Voter Models Be Improved?","slug":"can-likely-voter-models-be-improved","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/can-likely-voter-models-be-improved\/","is_active":false},{"id":101858,"title":"1. Polls and votes: The 2014 elections by the numbers","slug":"polls-and-votes-the-2014-elections-by-the-numbers","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/polls-and-votes-the-2014-elections-by-the-numbers\/","is_active":false},{"id":101843,"title":"2: Measuring the likelihood to vote","slug":"measuring-the-likelihood-to-vote","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/measuring-the-likelihood-to-vote\/","is_active":false},{"id":101827,"title":"3. Comparing the results of different likely voter models","slug":"comparing-the-results-of-different-likely-voter-models","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/comparing-the-results-of-different-likely-voter-models\/","is_active":false},{"id":101809,"title":"4. Conclusion","slug":"likely-voters-conclusion","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/likely-voters-conclusion\/","is_active":false},{"id":101793,"title":"Acknowledgements","slug":"likely-voters-acknowledgements","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/likely-voters-acknowledgements\/","is_active":false},{"id":101777,"title":"Methodology","slug":"methodology-239","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/methodology-239\/","is_active":false},{"id":101763,"title":"Appendix A: The Perry-Gallup measures","slug":"appendix-a-the-perry-gallup-measures","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-a-the-perry-gallup-measures\/","is_active":false},{"id":101753,"title":"Appendix B: The choice of a turnout measure","slug":"appendix-b-the-choice-of-a-turnout-measure","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-b-the-choice-of-a-turnout-measure\/","is_active":false},{"id":101872,"title":"Appendix C: Sensitivity to the turnout forecast","slug":"appendix-c-sensitivity-to-the-turnout-forecast","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-c-sensitivity-to-the-turnout-forecast\/","is_active":true}],"report_materials":"","report_pagination":{"current_post":{"id":101872,"title":"Appendix C: Sensitivity to the turnout forecast","slug":"appendix-c-sensitivity-to-the-turnout-forecast","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-c-sensitivity-to-the-turnout-forecast\/","is_active":true,"page_num":10},"next_post":null,"previous_post":{"id":101753,"title":"Appendix B: The choice of a turnout measure","slug":"appendix-b-the-choice-of-a-turnout-measure","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-b-the-choice-of-a-turnout-measure\/","is_active":false,"page_num":9},"pagination_items":[{"id":101878,"title":"Can Likely Voter Models Be Improved?","slug":"can-likely-voter-models-be-improved","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/can-likely-voter-models-be-improved\/","is_active":false,"page_num":1},{"id":101858,"title":"1. Polls and votes: The 2014 elections by the numbers","slug":"polls-and-votes-the-2014-elections-by-the-numbers","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/polls-and-votes-the-2014-elections-by-the-numbers\/","is_active":false,"page_num":2},{"id":101843,"title":"2: Measuring the likelihood to vote","slug":"measuring-the-likelihood-to-vote","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/measuring-the-likelihood-to-vote\/","is_active":false,"page_num":3},{"id":101827,"title":"3. Comparing the results of different likely voter models","slug":"comparing-the-results-of-different-likely-voter-models","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/comparing-the-results-of-different-likely-voter-models\/","is_active":false,"page_num":4},{"id":101809,"title":"4. Conclusion","slug":"likely-voters-conclusion","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/likely-voters-conclusion\/","is_active":false,"page_num":5},{"id":101793,"title":"Acknowledgements","slug":"likely-voters-acknowledgements","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/likely-voters-acknowledgements\/","is_active":false,"page_num":6},{"id":101777,"title":"Methodology","slug":"methodology-239","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/methodology-239\/","is_active":false,"page_num":7},{"id":101763,"title":"Appendix A: The Perry-Gallup measures","slug":"appendix-a-the-perry-gallup-measures","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-a-the-perry-gallup-measures\/","is_active":false,"page_num":8},{"id":101753,"title":"Appendix B: The choice of a turnout measure","slug":"appendix-b-the-choice-of-a-turnout-measure","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-b-the-choice-of-a-turnout-measure\/","is_active":false,"page_num":9},{"id":101872,"title":"Appendix C: Sensitivity to the turnout forecast","slug":"appendix-c-sensitivity-to-the-turnout-forecast","link":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/methods\/2016\/01\/07\/appendix-c-sensitivity-to-the-turnout-forecast\/","is_active":true,"page_num":10}]},"parent_info":{"parent_title":"Can Likely Voter Models Be Improved?","parent_id":101878},"materialsOrdered":[],"chaptersOrdered":[],"partsOrdered":[],"partsEnabled":false,"datacite_doi":"","prc_seo_data":{"title":"Can Likely U.S. Voter Models Be Improved?","description":"The candidate preferences of voters and nonvoters in 2014 were very different. This fact makes cutoff methods very sensitive to the chosen turnout threshold. Using the Perry-Gallup method, the forecast&hellip;","og_title":"Can Likely U.S. Voter Models Be Improved?","og_description":"The candidate preferences of voters and nonvoters in 2014 were very different. This fact makes cutoff methods very sensitive to the chosen turnout threshold. Using the Perry-Gallup method, the forecast&hellip;","schema_type":"Article","noindex":false,"canonical_url":"","primary_terms":[],"custom_schema":[],"og_image":0,"indexnow_submitted_at":null,"gsc_index_status":null},"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"relatedPostsOrdered":[],"bylinesOrdered":[],"acknowledgementsOrdered":[],"_links":{"self":[{"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/posts\/101872","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/users\/294"}],"replies":[{"embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/comments?post=101872"}],"version-history":[{"count":2,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/posts\/101872\/revisions"}],"predecessor-version":[{"id":134387,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/posts\/101872\/revisions\/134387"}],"wp:attachment":[{"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/media?parent=101872"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/categories?post=101872"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/tags?post=101872"},{"taxonomy":"bylines","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/bylines?post=101872"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/collection?post=101872"},{"taxonomy":"datasets","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/datasets?post=101872"},{"taxonomy":"level_of_effort","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/level_of_effort?post=101872"},{"taxonomy":"primary_audience","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/primary_audience?post=101872"},{"taxonomy":"information_type","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/information_type?post=101872"},{"taxonomy":"_post_visibility","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/_post_visibility?post=101872"},{"taxonomy":"formats","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/formats?post=101872"},{"taxonomy":"_fund_pool","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/_fund_pool?post=101872"},{"taxonomy":"languages","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/languages?post=101872"},{"taxonomy":"regions-countries","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/regions-countries?post=101872"},{"taxonomy":"research-teams","embeddable":true,"href":"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-json\/wp\/v2\/research-teams?post=101872"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}