[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-529690-105":59,"doc-detail-529690-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","determining-minimum-sample-sizes-for-multiple-regression-grade-prediction-equations-for-colleges-abstract","DETERMINING MINIMUM SAMPLE SIZES FOR MULTIPLE REGRESSION GRADE PREDICTION EQUATIONS FOR COLLEGES - ABSTRACT","","The American College Testing Program provides research services that help colleges predict the freshman grades of future students. This paper investigates the minimum sample size needed to calculate prediction equations for college freshman grade averages using multiple regression least-squares methods. Evidence from related studies indicates that eight-variable prediction equations built from representative samples of about 50 students can achieve accuracy comparable to equations based on substantially larger samples. The paper outlines how prediction error depends on the quality of weight estimates and examines approaches, including cross-college information sharing and centralized methods, to mitigate small-sample effects while defining practical MAE degradation thresholds.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/determining-minimum-sample-sizes-for-multiple-regression-grade-prediction-equations-for-colleges-abstract/529690/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/determining-minimum-sample-sizes-for-multiple-regression-grade-prediction-equations-for-colleges-abstract/529690.png","ImageObject",300,407,{"name":92,"@type":93},"awa","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-15","2026-10-09",true,{"@type":102,"interactionType":103,"userInteractionCount":34},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What is the document trying to determine?","Question",{"text":112,"@type":113},"It determines how small a college’s freshman class sample size can be for developing least-squares prediction equations without a significant drop in prediction accuracy.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"Why does sample size affect prediction accuracy?",{"text":117,"@type":113},"Because regression weights are estimated from data and their estimation error propagates into prediction error; small base samples can produce unacceptable accuracy.",{"name":119,"@type":110,"acceptedAnswer":120},"What sample size does the paper suggest is workable for eight-variable prediction equations?",{"text":121,"@type":113},"Results discussed indicate that representative samples of about 50 students can yield prediction accuracy almost as good as using larger samples.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},529690,1791529817,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":34,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":36,"preview_page_count":19,"preview_urls":145},3985747858093,"https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2","Determining Minimum Sample Sizes forMultiple Regression GradePrediction Equations for Colleges  \nRichard Sawyer  \nFebruary 1984  \nDETERMINING MINIMUM SAMPLE SIZES FOR MULTIPLEREGRESSION GRADE PREDICTION EQUATIONS FOR COLLEGES  \nPrepared by the Research and Development DivisionThe American Colege Testing Program  \nFor additional copies write:ACT PublicationsP.O.Box 168lowa City,lowa 52243  \n(Check or money order must accompany request.)Price:$1.00  \n⑥1984 by The American College Testing Program.All rights reserved.  \n# ABSTRACT\n\nThe American College Testing Program offers research services through which colleges can predictthe freshman grades of their future students.This paper describes research done to establish aminimum sample size requirement for calculating prediction equations for college freshman gradeaverage.Results from all the studies suggest that eight-variable prediction equations based onrepresentative samples of size 50 would have almost the same accuracy as prediction equationsbased on larger samples.  \nThe American Coillege Testing Program(ACT)offersresearch services through which colleges can predictthe freshman grades of their future students(TheAmerican College Testing Program,1983).The stu-dents'predicted grades are based on their ACT testscores in English,mathematics,social studies,andnatural sciences,and on their self-reported high schoolgrades in these four subject areas.The predictedgrades are calculated by weighting the test scores andhigh school grades in least-squares regression equa-tions that are specific to each college.1  \nThe weights in a college's prediction equation areusually calculated from data on all students in aprevious freshman class who took the ACT.Becausethese weights are estimates whose accuracy dependson the size of the base sample used to calculate them,and because error in estimating the weights propagateserror in prediction,the freshman class size affectsprediction error.It is possible,therefore,that weightscalculated from very small freshman classes could besubject to large sampling errors,resulting in predic-tions of unacceptable accuracy.  \nOne way to mitigate the effect of small sample sizes onprediction accuracy is to use information collaterallyfrom several colleges in constructing prediction equa-tions.Novick et al.(1972)further developed a Bayesianmodel due to Lindley(1970)in which this method wasused.Novick et al.calculated for m=22 junior collegesthe standard least-squares and the Bayesian \"m-group\"prediction equations for freshman grade average,usingthe four ACT test scores as predictors.The meannumber of students in the 22 colleges was approx-imately 246.Novick et al.then cross-validated theprediction equations against the following year's fresh-men at these colleges.They obtained an averagecross-validated Mean Absolute Error(MAE)of .58grade units for both the least-squares and the Bayesianm-group prediction methods.When the predictionequations were developed from 25%samples of thebase year freshman classes,the resulting mean cross-validated MAE was .61 grade units for the least-squares and.59 grade units for the Bayesian method.The results of Novick et al.suggest,therefore,thatfour-variable least-squares predictions for freshmanclasses wth as few as 50 students would not begrossly inaccurate.The results further suggest that theBayesian m-group method would yield more accurate  \n# DETERMINING MINIMUM SAMPLE SIZES FOR MULTIPLEREGRESSION GRADE PREDICTION EQUATIONS FOR COLLEGES\n\nRichard Sawyer  \npredictions than east-squares when sample sizes aresmaller than 50.Other centralized prediction methods,such as that due to Dempster,Rubin,and Tsutakawa(1981),also seem promising in this regard.  \nThe focus of this paper is on standard least-squarespredictions,since they are still the most extensivelyused predictions and are currently used by ACT.Thepurpose of the study is to determine for how small acollege least-squares prediction equations can bedeveloped without significant d","cbCaioE5GUI81cVl","https://ap.wps.com/l/cbCaioE5GUI81cVl","pdf",429421,16,"English","# ABSTRACT\n## Theoretical Considerations","[{\"question\":\"What is the document trying to determine?\",\"answer\":\"It determines how small a college’s freshman class sample size can be for developing least-squares prediction equations without a significant drop in prediction accuracy.\"},{\"question\":\"Why does sample size affect prediction accuracy?\",\"answer\":\"Because regression weights are estimated from data and their estimation error propagates into prediction error; small base samples can produce unacceptable accuracy.\"},{\"question\":\"What sample size does the paper suggest is workable for eight-variable prediction equations?\",\"answer\":\"Results discussed indicate that representative samples of about 50 students can yield prediction accuracy almost as good as using larger samples.\"}]","DETERMINING MINIMUM SAMPLE SIZES FOR MULTIPLE REGRESSION GRADE PREDICTION EQUATIONS FOR COLLEGES - ABSTRACT | PDF",1791521756,[146,150,154,158],{"page_no":8,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/983865c591769062c814772ee3cf2574.webp",983,1271,{"page_no":14,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/88be9d50cd7f7f7ae7fdaaca73e049f1.webp",982,1266,{"page_no":81,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/21b1bfa3d9a96a60eaa3a3b7a51ec4dc.webp",981,1311,{"page_no":19,"url":159,"width":160,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/284cac3755ffdc7f25a51cdc6e52621a.webp",980]