Designing Your Extraction Codebook | HubMeta Tutorial #14 — Transcript
Full transcript
- 0:15All right, welcome back. So the big step
- 0:20that now we have to set up is setting up
- 0:24our project code book. Think of a
- 0:26project code book like a template that
- 0:29you create like an Excel file with all
- 0:32the columns and all the explanations and
- 0:35you will print this. You will uh have it
- 0:38in front of you so that each one of the
- 0:41papers that you are looking at you will
- 0:43know that this is the data these are the
- 0:46fields this is the data that I will
- 0:48extract from this paper. Right? What you
- 0:53actually extract from each paper depends
- 0:56very very much on your project, the
- 0:59analysis you want to do, but basically
- 1:02what questions are you asking of all of
- 1:06your articles, right? Are you asking
- 1:08about the theories they have used? Are
- 1:10you asking about the sample size, the
- 1:12the context like the locations?
- 1:15um um you know um the intervention that
- 1:20was done, you know, all of that. And of
- 1:23course the results of the findings like
- 1:27if it is a quantitative study, we
- 1:29usually look at things such as the
- 1:31effect size or a correlation
- 1:34uh you know something that can be
- 1:36converted to a um to an effect size. Or
- 1:39if it is a qualitative study, we might
- 1:42have many many different questions like
- 1:45what theories were used like how data
- 1:47was collected, you know, example codes
- 1:51from our participants,
- 1:54uh what what were the findings, what
- 1:57context,
- 1:58you name it. The best person to set up
- 2:02this code book is you as the researcher.
- 2:07Usually uh this step is where you have
- 2:10to sit with yourself and think you know
- 2:13for a long time what do I want to know
- 2:16and in that you can actually use AI to
- 2:20you know um to set up the codebook to
- 2:24give you some guidance as to what you
- 2:26need to include. Now, of course, you can
- 2:28ask AI to help you with setting up this
- 2:30codebook, but really this is the time
- 2:33that you want to really think as a
- 2:35researcher. What do I want to know? And
- 2:39when you had some idea, you can start
- 2:42creating new templates here. Inside this
- 2:46you can give it a name like for example
- 2:49sample characteristics
- 2:51and then you know add a description and
- 2:54then add field by field like for example
- 2:57sample size. Is this a number a you know
- 3:01a text a drop-down with preset values?
- 3:06Uh like for example are you trying to
- 3:08capture different countries? If it is a
- 3:12drop-down, you have the option to pick
- 3:14for example countries. These are
- 3:15prepopulated for you. That's one way.
- 3:19But the other way which is you know
- 3:22easier to use is to use our AI generate
- 3:26function. Here you can provide a
- 3:30complete explanation and AI will help
- 3:33you generate the fields that you want.
- 3:35like I can basically explain uh sample
- 3:39characteristics
- 3:41uh for firm level studies like I don't
- 3:47know things such as industry country
- 3:51uh sample size number of firms and uh uh
- 3:57you know a few things like that and when
- 3:59I uh say generate this will this is the
- 4:05specifically created AI that we have in
- 4:08the background that will generate these
- 4:10fields. So it says uh fields that will
- 4:14be captured in this template include
- 4:16industry, country, sample size, number
- 4:18of firms for each of them a definition,
- 4:21a type and if you were happy with this,
- 4:24you can say use this. You can of course
- 4:27before saving this template you can edit
- 4:30uh all of these fills and then you can
- 4:34uh
- 4:36um you can just go ahead and save it
- 4:39because this is a live project. I will
- 4:41just add some number at the start of the
- 4:43name. So I recognize that this was just
- 4:46an example template. But you can see all
- 4:49of these are templates that I have
- 4:51created uh from before. Like this one is
- 4:53for example just bibliographic
- 4:55information uh of the article. In this
- 4:58one you know the quality level uh of the
- 5:01article. This one you know the dates it
- 5:05was collected. So this one what
- 5:09measurements were used in that study but
- 5:11whatever it is. So we will create
- 5:13multiple data and then all of these will
- 5:16be captured. In addition, when you have
- 5:19system tables,
- 5:22this will allow you to include
- 5:25correlation tables and regression
- 5:28tables. So here uh in a new project, all
- 5:31you got to do is to add correlation
- 5:33table and regression table. Then
- 5:36correlation and regression will be added
- 5:38to the things that will be captured for
- 5:40this study. for your correlation matrix.
- 5:43We have a lot of options of all the
- 5:46columns in addition to the correlation
- 5:49that can be extracted. So carefully
- 5:52review this. You might not want all of
- 5:55this. This has been added after years of
- 5:58doing meta analysis with us. Like these
- 6:01are things that we needed in many
- 6:03different uh projects or you might need
- 6:06more. So here you can add more column
- 6:08names and add a definition and
- 6:11correlation table will also capture
- 6:14that. So now
- 6:17we have set up our code book. We
- 6:20reviewed what library and capture is. In
- 6:24the next video we will talk about how to
- 6:27use AI to extract the data that we need
- 6:31for our papers.
About this transcript
This page contains the full transcript of Designing Your Extraction Codebook | HubMeta Tutorial #14 by HubMeta, generated from the public captions YouTube serves with the video. The transcript has 897 words across 126 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.