[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-665385-105":3,"detail-sidebar-cat-0-en-105":80,"doc-detail-665385-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":73,"head_meta":75,"extra_data":77,"updated_unix":79},105,"en","good-and-bad-controls-an-introduction-to-causal-diagrams","Good and Bad Controls - An Introduction to Causal Diagrams","","A concise introduction to causal diagrams framed around the theme of “good” versus “bad” controls in observational data analysis. The material motivates why causal questions require external contextual knowledge that cannot be inferred from data alone, then contrasts correlation with causation. It reviews traditional limitations in encoding causal knowledge and argues for a modern causal inference approach that begins with explicit causal aims, highlighting how inappropriate conditioning leads to misleading estimates.",{"@graph":14,"@context":72},[15,34,55],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/good-and-bad-controls-an-introduction-to-causal-diagrams/665385/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/good-and-bad-controls-an-introduction-to-causal-diagrams/665385.png","ImageObject",300,407,{"name":42,"@type":43},"cksankcs","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-30","2026-10-29",true,{"@type":52,"interactionType":53,"userInteractionCount":33},"InteractionCounter",{"@type":54},"ViewAction",{"@type":56,"mainEntity":57},"FAQPage",[58,64,68],{"name":59,"@type":60,"acceptedAnswer":61},"Why do causal questions require external knowledge?","Question",{"text":62,"@type":63},"Because the decision to control for a third variable depends on context that is not knowable from the data itself.","Answer",{"name":65,"@type":60,"acceptedAnswer":66},"What is the limitation of estimating causal effects from observational data traditionally?",{"text":67,"@type":63},"There has been no formal way to encode causal knowledge, leading to poor and inconsistent records of causal estimation from observational studies.",{"name":69,"@type":60,"acceptedAnswer":70},"How should observational findings be communicated regarding causation?",{"text":71,"@type":63},"Causal language is often discouraged for observational interpretation, with studies relying on terms like risk factor, association, predictor, or correlation instead of explicit causation.","https://schema.org",{"og:url":32,"og:type":74,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":76,"canonical":32},"index,follow",{"doc_id":78,"site_id":7},665385,1793302013,{"code":4,"msg":81,"data":82},"success",[83,87,91,95,100,105,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":84,"show_sort_weight":85,"slug":86},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":88,"show_sort_weight":89,"slug":90},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":92,"show_sort_weight":93,"slug":94},"Exam",70,"exam",{"id":96,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},5,"Comic",60,"comic",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},6,"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},7,"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},8,30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":96,"slug":129},19,"General","general",{"code":4,"msg":81,"data":131},{"doc_id":78,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":111,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":33,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":79,"read_time":144},3985747882755,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","GOOD AND BAD CONTROLS:  \nAN INTRODUCTION TO CAUSAL DIAGRAMS  \nPeter WG Tennant PhD  \nAssociate Professor of Health Data Science University of Leeds  \n[tinyurl.com/pwgtennant](tinyurl.com/pwgtennant)  @PWGTennant  \nTO CONTROL OR NOT TO CONTROL?  \nOverall  \n\n| \u003Cbr>Men |  |  | Women |\n| --- | --- | --- | --- |\n| N | Mean grant income | N | Mean grant income |\n| 200 | £54,338 | 300 | £43,700 |\n\n@PWGTennant  \nTO CONTROL OR NOT TO CONTROL?  \nOverall  \nLight teaching & citizenship Heavy teaching & citizenship  \n\n|  | Men |  | Women |\n| --- | --- | --- | --- |\n| N | Mean grant income | N | Mean grant income |\n| 200 | £54,338 | 300 | £43,700 |\n| 115 | £69,000 | 80 | £69,000 |\n| 85 | £34,500 | 220 | £34,500 |\n\n@PWGTennant  \nTO CONTROL OR NOT TO CONTROL?  \nAverage annual wage  \nOverall  \n\n| \u003Cbr>Men |  |  | Women |\n| --- | --- | --- | --- |\n| N | Mean annual wage | N | Mean annual wage |\n| 200 | £54,338 | 300 | £43,700 |\n\n@PWGTennant  \nTO CONTROL OR NOT TO CONTROL?  \nAverage annual wage  \nSenior academic  \nJunior academic  \nOverall  \nSenior academic  \nJunior Academic  \n\n|  | Men |  | Women |\n| --- | --- | --- | --- |\n| N | Mean annual wage | N | Mean annual wage |\n| 200 | £54,338 | 300 | £43,700 |\n| 115 | £69,000 | 80 | £69,000 |\n| 85 | £34,500 | 220 | £34,500 |\n\n@PWGTennant  \nCAUSAL QUESTIONS REQUIRE EXTERNAL KNOWLEDGE  \n§ In the previous examples, we (implicitly) wanted to know:  \n§ Causal effect of male vs female on average yearly grant income  \n§ Causal effect of male vs female on average pay  \n§ The data were identical in both scenarios  \n§ But the decision of whether or not to control for a third variable differed  \n§ We used external knowledge of the context – not knowable from the data itself-to reason whether it was appropriate to condition on a third variable  \n @PWGTennant  \nESTIMATING CAUSAL EFFECTS FROM OBSERVATIONAL DATA  \n§ Traditionally, we had no formal way to encode this knowledge  \n§ Was done on ad-hoc basis (if at all!)  \n§ Consequently, record of estimating causal effects from observational data is poor  \nþ  \n @PWGTennant  \ný  \nCORRELATION ≠ CAUSATION  \n§ Discouraged from using causal language when interpreting observational data  \n§ Observational studies routinely rated as‘low quality’  \nGuyatt G, Oxman AD, Akl EA, et al. J Clin Epidemiol 2011;64:383–94  \n§ Researchers have relied on euphemisms  \n“Risk factor”“Link”  \n“Correlation”  \n“Predictor”  \n“Association”  \n @PWGTennant  \nEMBRACING CAUSAL INFERENCE  \n§ But euphemisms = unclear aims = inappropriate methods = misleading estimates  \n§ A modern causal inference approach starts with being about your causal aims!  \nAmerican Journal of Public Health-May 2018  \n @PWGTennant","cbCailApqfjxIzta","https://ap.wps.com/l/cbCailApqfjxIzta","pdf",5093250,48,"English","# Good and Bad Controls - An Introduction to Causal Diagrams\n## To Control or Not to Control? - Overall\n## To Control or Not to Control? - Teaching and Citizenship\n## To Control or Not to Control? - Average Annual Wage\n## Causal Questions Require External Knowledge\n## Estimating Causal Effects from Observational Data\n## Correlation ≠ Causation\n## Embracing Causal Inference","[{\"question\":\"Why do causal questions require external knowledge?\",\"answer\":\"Because the decision to control for a third variable depends on context that is not knowable from the data itself.\"},{\"question\":\"What is the limitation of estimating causal effects from observational data traditionally?\",\"answer\":\"There has been no formal way to encode causal knowledge, leading to poor and inconsistent records of causal estimation from observational studies.\"},{\"question\":\"How should observational findings be communicated regarding causation?\",\"answer\":\"Causal language is often discouraged for observational interpretation, with studies relying on terms like risk factor, association, predictor, or correlation instead of explicit causation.\"}]","Good and Bad Controls - An Introduction to Causal Diagrams | PDF",121]