Invited commentary: Comparing the independent segments procedure with group sequential designs

Invited commentary: Comparing the independent segments procedure with group sequential designs. Lakens, D. Psychological Methods, 26(4):498–500, August, 2021. Num Pages: 498-500 Place: Washington, US Publisher: American Psychological Association (US)

Paper doi abstract bibtex

Psychological science would become more efficient if researchers implemented sequential designs where feasible. Miller and Ulrich (2020) propose an independent segments procedure where data can be analyzed at a prespecified number of equally spaced looks while controlling the Type I error rate. Such procedures already exist in the sequential analysis literature, and in this commentary, I reflect on whether psychologists should choose to adopt these existing procedures instead. I believe limitations in the independent segments procedure make it relatively unattractive. Being forced to stop for futility based on a bound not chosen to control Type II errors, or reject a smallest effect size of interest in an equivalence test, limits the inferences one can make. Having to use a prespecified number of equally spaced looks is logistically inconvenient. And not having the flexibility to choose α and β spending functions limits the possibility to design efficient studies based on the goal and limitations of the researcher. Recent software packages such as rpact (Wassmer & Pahlke, 2019) make sequential designs equally easy to perform as the independent segments procedure. While learning new statistical methods always takes time, I believe psychological scientists should start on a path that will not limit them in the flexibility and inferences their statistical procedure provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved) (Source: journal abstract) \textlessstrong xmlns:lang="en"\textgreaterTranslational Abstract—Psychological science will become more efficient if researchers analyze data repeatedly as it comes in. This allows scientists to decide if the data collection can be stopped early because the data already answers the research question. These techniques, which limit the probability of drawing erroneous conclusions, are called ‘sequential analysis’. Miller and Ulrich (2020) propose the independent segments procedure with the goal to provide researchers with an easy to use approach to perform sequential analysis. In this comment I point out some limitations compared to a sequential analysis technique that is more established in other fields, which is called group sequential designs. I believe psychologists will benefit from the greater flexibility that group sequential designs provide compared to the independent segments procedure. If psychologists spend time learning a new statistical technique, they should start on a path that will not limit them in the flexibility the technique provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved)

@article{lakens_invited_2021,
	title = {Invited commentary: {Comparing} the independent segments procedure with group sequential designs},
	volume = {26},
	copyright = {© 2021, American Psychological Association},
	issn = {1082-989X},
	shorttitle = {Invited commentary},
	url = {https://www.proquest.com/docview/2587210431/abstract/798884655D9B4B2FPQ/1},
	doi = {http://dx.doi.org/10.1037/met0000400},
	abstract = {Psychological science would become more efficient if researchers implemented sequential designs where feasible. Miller and Ulrich (2020) propose an independent segments procedure where data can be analyzed at a prespecified number of equally spaced looks while controlling the Type I error rate. Such procedures already exist in the sequential analysis literature, and in this commentary, I reflect on whether psychologists should choose to adopt these existing procedures instead. I believe limitations in the independent segments procedure make it relatively unattractive. Being forced to stop for futility based on a bound not chosen to control Type II errors, or reject a smallest effect size of interest in an equivalence test, limits the inferences one can make. Having to use a prespecified number of equally spaced looks is logistically inconvenient. And not having the flexibility to choose α and β spending functions limits the possibility to design efficient studies based on the goal and limitations of the researcher. Recent software packages such as rpact (Wassmer \& Pahlke, 2019) make sequential designs equally easy to perform as the independent segments procedure. While learning new statistical methods always takes time, I believe psychological scientists should start on a path that will not limit them in the flexibility and inferences their statistical procedure provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved) (Source: journal abstract)
{\textless}strong xmlns:lang="en"{\textgreater}Translational Abstract—Psychological science will become more efficient if researchers analyze data repeatedly as it comes in. This allows scientists to decide if the data collection can be stopped early because the data already answers the research question. These techniques, which limit the probability of drawing erroneous conclusions, are called ‘sequential analysis’. Miller and Ulrich (2020) propose the independent segments procedure with the goal to provide researchers with an easy to use approach to perform sequential analysis. In this comment I point out some limitations compared to a sequential analysis technique that is more established in other fields, which is called group sequential designs. I believe psychologists will benefit from the greater flexibility that group sequential designs provide compared to the independent segments procedure. If psychologists spend time learning a new statistical technique, they should start on a path that will not limit them in the flexibility the technique provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved)},
	language = {English},
	number = {4},
	urldate = {2021-11-15},
	journal = {Psychological Methods},
	author = {Lakens, Daniël},
	month = aug,
	year = {2021},
	note = {Num Pages: 498-500
Place: Washington, US
Publisher: American Psychological Association
(US)},
	keywords = {Behavioral Sciences (major), Hypothesis Testing (major), Learning (major), Methodology, Sampling (Experimental), Statistical Tests, Type I Errors (major), Type II Errors (major)},
	pages = {498--500},
}

Downloads: 0

{"_id":"fbgYKJdk3Zg6h7pko","bibbaseid":"lakens-invitedcommentarycomparingtheindependentsegmentsprocedurewithgroupsequentialdesigns-2021","author_short":["Lakens, D."],"bibdata":{"bibtype":"article","type":"article","title":"Invited commentary: Comparing the independent segments procedure with group sequential designs","volume":"26","copyright":"© 2021, American Psychological Association","issn":"1082-989X","shorttitle":"Invited commentary","url":"https://www.proquest.com/docview/2587210431/abstract/798884655D9B4B2FPQ/1","doi":"http://dx.doi.org/10.1037/met0000400","abstract":"Psychological science would become more efficient if researchers implemented sequential designs where feasible. Miller and Ulrich (2020) propose an independent segments procedure where data can be analyzed at a prespecified number of equally spaced looks while controlling the Type I error rate. Such procedures already exist in the sequential analysis literature, and in this commentary, I reflect on whether psychologists should choose to adopt these existing procedures instead. I believe limitations in the independent segments procedure make it relatively unattractive. Being forced to stop for futility based on a bound not chosen to control Type II errors, or reject a smallest effect size of interest in an equivalence test, limits the inferences one can make. Having to use a prespecified number of equally spaced looks is logistically inconvenient. And not having the flexibility to choose α and β spending functions limits the possibility to design efficient studies based on the goal and limitations of the researcher. Recent software packages such as rpact (Wassmer & Pahlke, 2019) make sequential designs equally easy to perform as the independent segments procedure. While learning new statistical methods always takes time, I believe psychological scientists should start on a path that will not limit them in the flexibility and inferences their statistical procedure provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved) (Source: journal abstract) \\textlessstrong xmlns:lang=\"en\"\\textgreaterTranslational Abstract—Psychological science will become more efficient if researchers analyze data repeatedly as it comes in. This allows scientists to decide if the data collection can be stopped early because the data already answers the research question. These techniques, which limit the probability of drawing erroneous conclusions, are called ‘sequential analysis’. Miller and Ulrich (2020) propose the independent segments procedure with the goal to provide researchers with an easy to use approach to perform sequential analysis. In this comment I point out some limitations compared to a sequential analysis technique that is more established in other fields, which is called group sequential designs. I believe psychologists will benefit from the greater flexibility that group sequential designs provide compared to the independent segments procedure. If psychologists spend time learning a new statistical technique, they should start on a path that will not limit them in the flexibility the technique provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved)","language":"English","number":"4","urldate":"2021-11-15","journal":"Psychological Methods","author":[{"propositions":[],"lastnames":["Lakens"],"firstnames":["Daniël"],"suffixes":[]}],"month":"August","year":"2021","note":"Num Pages: 498-500 Place: Washington, US Publisher: American Psychological Association (US)","keywords":"Behavioral Sciences (major), Hypothesis Testing (major), Learning (major), Methodology, Sampling (Experimental), Statistical Tests, Type I Errors (major), Type II Errors (major)","pages":"498–500","bibtex":"@article{lakens_invited_2021,\n\ttitle = {Invited commentary: {Comparing} the independent segments procedure with group sequential designs},\n\tvolume = {26},\n\tcopyright = {© 2021, American Psychological Association},\n\tissn = {1082-989X},\n\tshorttitle = {Invited commentary},\n\turl = {https://www.proquest.com/docview/2587210431/abstract/798884655D9B4B2FPQ/1},\n\tdoi = {http://dx.doi.org/10.1037/met0000400},\n\tabstract = {Psychological science would become more efficient if researchers implemented sequential designs where feasible. Miller and Ulrich (2020) propose an independent segments procedure where data can be analyzed at a prespecified number of equally spaced looks while controlling the Type I error rate. Such procedures already exist in the sequential analysis literature, and in this commentary, I reflect on whether psychologists should choose to adopt these existing procedures instead. I believe limitations in the independent segments procedure make it relatively unattractive. Being forced to stop for futility based on a bound not chosen to control Type II errors, or reject a smallest effect size of interest in an equivalence test, limits the inferences one can make. Having to use a prespecified number of equally spaced looks is logistically inconvenient. And not having the flexibility to choose α and β spending functions limits the possibility to design efficient studies based on the goal and limitations of the researcher. Recent software packages such as rpact (Wassmer \\& Pahlke, 2019) make sequential designs equally easy to perform as the independent segments procedure. While learning new statistical methods always takes time, I believe psychological scientists should start on a path that will not limit them in the flexibility and inferences their statistical procedure provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved) (Source: journal abstract)\n{\\textless}strong xmlns:lang=\"en\"{\\textgreater}Translational Abstract—Psychological science will become more efficient if researchers analyze data repeatedly as it comes in. This allows scientists to decide if the data collection can be stopped early because the data already answers the research question. These techniques, which limit the probability of drawing erroneous conclusions, are called ‘sequential analysis’. Miller and Ulrich (2020) propose the independent segments procedure with the goal to provide researchers with an easy to use approach to perform sequential analysis. In this comment I point out some limitations compared to a sequential analysis technique that is more established in other fields, which is called group sequential designs. I believe psychologists will benefit from the greater flexibility that group sequential designs provide compared to the independent segments procedure. If psychologists spend time learning a new statistical technique, they should start on a path that will not limit them in the flexibility the technique provides. (PsycInfo Database Record (c) 2021 APA, all rights reserved)},\n\tlanguage = {English},\n\tnumber = {4},\n\turldate = {2021-11-15},\n\tjournal = {Psychological Methods},\n\tauthor = {Lakens, Daniël},\n\tmonth = aug,\n\tyear = {2021},\n\tnote = {Num Pages: 498-500\nPlace: Washington, US\nPublisher: American Psychological Association\n(US)},\n\tkeywords = {Behavioral Sciences (major), Hypothesis Testing (major), Learning (major), Methodology, Sampling (Experimental), Statistical Tests, Type I Errors (major), Type II Errors (major)},\n\tpages = {498--500},\n}\n\n","author_short":["Lakens, D."],"key":"lakens_invited_2021","id":"lakens_invited_2021","bibbaseid":"lakens-invitedcommentarycomparingtheindependentsegmentsprocedurewithgroupsequentialdesigns-2021","role":"author","urls":{"Paper":"https://www.proquest.com/docview/2587210431/abstract/798884655D9B4B2FPQ/1"},"keyword":["Behavioral Sciences (major)","Hypothesis Testing (major)","Learning (major)","Methodology","Sampling (Experimental)","Statistical Tests","Type I Errors (major)","Type II Errors (major)"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://api.zotero.org/users/371621/collections/RZBZZXIB/items?key=EqaBcWShhjLdAyw7hySjsqNi&format=bibtex&limit=100","dataSources":["oM2qhrB64zmZJ375M"],"keywords":["behavioral sciences (major)","hypothesis testing (major)","learning (major)","methodology","sampling (experimental)","statistical tests","type i errors (major)","type ii errors (major)"],"search_terms":["invited","commentary","comparing","independent","segments","procedure","group","sequential","designs","lakens"],"title":"Invited commentary: Comparing the independent segments procedure with group sequential designs","year":2021}