Am I right in thinking that Scrivener is not for research


I had a look at Scrivener due to the appeal it has when it comes to structuring text and working with material across a larger text. However, I cannot find a mention anywhere (or maybe I was looking in the wrong places) of Figures, Tables. Also, I can’t find any information about referencing to figures and tables. Am I right?

So, maybe Scrivener is not so great for empirical research purposes (despite the APA style)?


I would expect a larger response shortly from somebody who knows those features better than me but Tables are certainly supported. I use them myself.

As far as figures? well if you are talking about generating images from equations etc. then I’m not sure that’s a feature but it does support images created in other things.

It also has some integration with Bibliography and Citation managers but I haven’t tried that. I would suggest keep digging/playing with it. There’s a lot that’s not immediately apparent when you are still getting to grips with it and in the meantime one of the folks that really know the package may come along… :slight_smile:

I think it is fair to say that Scrivener was originally conceived for the purpose of writing fiction, and some of its features reflect that, but plenty of people use it successfully for writing non-fiction. In fact, I would go so far as to say that for certain types of non-fiction, like history, it would be difficult to find a better tool. I speak as someone who wrote a 560 page history book in Word 5 on a PowerBook 180, and have just finished an 80,000 word thesis on history and psychology, much of which was written in Scrivener 1.5. I have had some dissatisfactions about footnotes and endnotes in Scrivener, but I understand that these have been addressed, and that an imminent version will go a long way to solving the problem.

I presently use Bookends as my bibliographic manager, and I keep pdfs of articles (mainly on psychology, but also on history) in a folder that is managed by Bookends. I don’t usually use figures, but if I did I would probably use Bookends to manage those, as well, and do any editing in an external editor, like Pixelmator, or OmniGraffle. I have developed the habit of using Devonthink Pro Office as the main repository for my “research materials” – in other words, notes, quotations, clippings from web pages, and so forth. I also keep the Bookends folder of pdfs indexed by Devonthink Pro, so that all the information from those articles is immediately available in Devonthink. I use Devonthink as my main research depository and organisational tool because my projects can get so large that they become difficult to manage in Scrivener (the last project had over 3.1 million words, 493 pdfs, 918 rtfs, and 622 images). Scrivener gets a bit cramped if you try to cram that much into it – it’s not what it was designed for, I would say. But when it comes to the actual writing, there is nothing better than Scrivener for organising the text. Where you refer to articles or other works, you insert temporary citations from Bookends, you can put in figures if you want (I now remember I did actually use one), and when the text is complete, you compile the text, and do the final formatting, including bibliography, table of contents, in a word-processor (I used Nisus Writer Pro).

So you can definitely use Scrivener for certain sorts of research – I’ve done it.

Best wishes, Martin.

Edit: I should mention that there are various ways of linking between Scrivener and material in Bookends and Devonthink, so that you can refer to the material while writing in Scrivener. Or you can, of course, import material into Scrivener when you need it, and delete it later if it gets in the way, knowing that the original copy will still be available in Bookends or Devonthink. And I should also say that there are other programs, like BibDesk, Sente, EagleFiler, and so forth, that do similar things to Bookends and Devonthink. It’s a matter of choosing what one can afford, and what suits one’s methods.

Second edit: perhaps I should clarify what I mean by keeping my pdfs “indexed” by Devonthink. You can either import material into Devonthink (so that the file sits inside DT’s folders), or you can keep material in its own folder, and have DT “index” it – which means in the case of files that have text (pdfs, rtfs, docs, etc) that all of the text gets read into DT, and can be analysed and searched using DT’s (very sophisticated) AI and search functions. I do this with pdfs of articles because Bookends needs to have them in its own folder so it can keep track of them, rename them, etc.

It still surprises me that the research community has never come up with an “integrated research tool” that combines everything in one piece of software, but perhaps our computers are not yet powerful enough to cope with what would be a very large and complex program. And using three separate programs is not too bad, given that they will talk to each other reasonably politely – except that I now realise I probably have to learn Tinderbox, as well. Help.

Depends what you mean by research - it covers everything from finding interviewees for local radio to the Manhattan Project.

Scrivener’s good for pulling together information for drama and fiction, and by the sounds of it pretty good at humanities style research. But stuff with numbers? Probably not.

Actually, I have found Scrivener great for writing empirical research articles. Certainly much better than a word processor. For one thing, while writing the discussion section, I can have the methods and results sections up in either quick reference windows or in the other pane. Also the ability to easily restructure parts of my argument (most useful in the introduction or discussion sections). There is not an automatic way to number figures or tables; for example, insert a placeholder in the text referring to the figure, and having the program generate a figure number based on where it appears in the text. But I know of no software that does that.

(edit) There actually is a way to insert placeholders and number figures and tables dynamically. See here:

No, you cannot draw figures in Scrivener, but there are much better programs than a writing program for that. I use OmniGraffle which can generate wonderful figures quite easily and flexibly. These are then exported into Scrivener and placed after the references in a separate document. I recently did this for a comprehensive review article I wrote, and the results were excellent. I have heard Adobe Illustrator is excellent, but has a steep learning curve.

Tables in Scrivener are quite rudimentary, but adequate for tables of means and standard deviations and effect sizes, for example. I think the main reason is that the tables need to render properly in the standard RTF file format on export, so the features of tables by necessity are restricted to what’s available through RTF. The formatting of the tables (rules and lines, for example) is better done in the word processor, just as is any complex text formatting. Remember Scrivener is a tool for thinking and writing, whereas the word processor’s job it to make it look pretty. My experience is that RTF export of tables is best handled by Word, unfortunately, rather than Pages.

One of the reasons I love Scrivener for my academic writing is the ability to have the notes I’ve taken from my articles in one pane, while having the text of the document in the other pane, so I can refer back and forth easily. That’s great for writing the introduction section for example. I can also print the SPSS output, and place the PDF into Scrivener’s Research folder, and refer to that in one pane while writing the results section in the other.

I use Sente to manage my references, including using it to take quotes and notes while reading (either on iPad or on my Mac), and then I export those notes to Scrivener, and place them in the Research folder for use while writing. I wrote an Applescript to do the export, BTW. When I reference an article , I place a citation tag inline (or drag and drop/copy and paste from Sente), and after the Scrivener text is rendered to the word processor, I run the WP text through Sente to generate the bibliography.

Attached is a screen shot of my recent writing project including the inclusion of a figure. Also a screen shot of how I use notes taken from an article when writing about the article.

Sorry for the delay,

My 30 days trial expired, and I have to say that I was sure that Scrivener was not for me at that point. Indeed, I did not even check whether I got any replies on this post… Sorry! :blush:

BUT!.. I got back to reading parts of the Scrivener manual for some reason, and found all this nice information concerning automatic numbering codes and stuff. Now I have to say that I cannot see what would be missing in Scrivener for my purposes. Well, generation of reference lists from Papers2 right within Scrivener perhaps… That is the only thing, I think.

Just excellent software this!


Glad you saw the light! :slight_smile: Seriously, there are different products that work with the way people think differently, so for some Scrivener might be the right tool, whereas for others a word processor is better. It works for me.

As a follow-up to the numbering of figures, etc. I altered how I handle them. In the attached screen shot, you’ll see I give the figure document the name “Figure <$n:figure:Delegit>” and in the caption for the figure refer to <$title> (in italics per APA style) to have the figure’s name inserted. That way, if I move the figure around it will be dynamically renumbered and labeled. I also refer to the figure in the text as “(Figure <$n:figure:Delegit>)” so it picks up the number from the dynamic numbering.