tag:blogger.com,1999:blog-5841910768079015534.post8199997904172423929..comments2024-03-18T08:28:01.624+00:00Comments on BishopBlog: Reproducible practices are the future for early career researchersdeevybeehttp://www.blogger.com/profile/15118040887173718391noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-5841910768079015534.post-28741849206716870442017-05-10T14:57:04.994+01:002017-05-10T14:57:04.994+01:00Good talk by Markowetz.
I'm a firm believer i...Good talk by Markowetz.<br /><br />I'm a firm believer in using Latex (well actually Lyx) with knitr and R for writing a paper.<br /><br />That way one does not forget to change a p-value or n-size if the data changes slightly. <br /><br /><br />jrkrideauhttps://www.blogger.com/profile/04869979887929067657noreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-56083145518479898712017-05-03T01:48:17.081+01:002017-05-03T01:48:17.081+01:00Oddly enough, there is a great cover version of th...Oddly enough, there is a great cover version of that Bob Dylan song with lyrics about reproducible science:<br /><br />https://www.youtube.com/watch?v=2ze-_jA1X94&feature=youtu.be <br /><br />Or search for "The statistics, they are a changin' "Daniel Feuerriegelnoreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-54950741938508772852017-05-02T08:54:34.248+01:002017-05-02T08:54:34.248+01:00"... sticking to the status quo ..." nai..."... sticking to the status quo ..." nails a large part of the problem. I think this also applies to undergraduate teaching: at least in my experience, 'Methods' and 'Data Analysis' courses are seldom about methods or data analysis, but instead show students how to enter numbers into SPSS and scan the output. Why? Because it's easy, and we've always done it that way ...<br /><br />GeoffGeoff Hammondhttps://www.blogger.com/profile/06787242437033453489noreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-63535946577387376502017-05-02T06:39:15.097+01:002017-05-02T06:39:15.097+01:00While using Git is a huge step in the right direct...While using Git is a huge step in the right direction, it isn't really indicative of the data integrity being maintained, Unless the entire data analysis process was scrupulously managed in a git repository, not just the 'final' datasets and script.<br /><br />Granted, your move to require that from dissertees will allow this conversation to start, and will help in the creation of a generation of data-integrity aware scientists, from within the less computationally literate population.<br /><br />Kudos to you and yours ! Naveh Shetrithttps://twitter.com/navshnoreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-29172092094048722092017-05-01T20:19:13.118+01:002017-05-01T20:19:13.118+01:00I suggested we include 'data integrity' as...I suggested we include 'data integrity' as a marked component of our final year students' dissertation module, but hit a brick wall. No-one saw the need - "they have to give us their data on a USB stick, isn't that enough?". Part of the problem is that most established academics don't know how to make their data or analyses open, so cannot train their students to do so. Next year my dissertees will be putting data and R scripts onto github. One dissertation supervisor at a time...jon.mayhttps://www.blogger.com/profile/02785548135187338314noreply@blogger.comtag:blogger.com,1999:blog-5841910768079015534.post-52468672023679301582017-05-01T19:46:10.728+01:002017-05-01T19:46:10.728+01:00Hear hear! Later in the song:
Come mothers and f...Hear hear! Later in the song:<br /><br />Come mothers and fathers<br />Throughout the land<br />And don't criticize<br />What you can't understand<br />Your sons and your daughters<br />Are beyond your command<br />Your old road is rapidly aging<br />Please get out of the new one if you can't lend your hand<br />Cause the times they are a-changing.<br /><br />SteveAnonymoushttps://www.blogger.com/profile/07485890516098106807noreply@blogger.com