how have you optimized your reading habits?

[spectastic]

Can we share ideas on ways we can get better at reading effectively?

I use mendeley, try to do lit review style bullet point summaries of the papers I read for later

I need to work on getting better at setting benchmarks for myself, otherwise, I'm just stuck in an ether of "wtf did I just do today?"

how about you?

[travelgirl125]

I would love to hear people's ideas!  I am a terribly slow reader and feel the need to understand every word I'm reading.  Everyone tells me I need to skim, but I feel like I need to actually learn how to do that!

Unfortunately, after week 1 I realized I would not be able to complete everything I had to do and read every single word.  So now I stick to reading abstracts, introductions, and conclusions.  I take notes and write a 1 sentence summary at the end of each reading.

[spectastic]

I do the same thing. I read the abstract and conclusion first to decide if this is even a paper worth reading. intro is sometimes helpful for leaning more info, but a lot of times, I'm reading the same things over and over

[Neist]

I've been using OCR'ed PDFs and text-to-speech while simultaneously reading, so I get visual and audible reinforcement. It's much faster, but it's still mentally taxing; it requires a fair amount of concentration.

Yesterday I burned through 350+ pages of reading. Wave of the future.  

[rising_star]

Set aside specific time for reading. For me, that's often the morning because I write better in the afternoon. But I also limit it to keep my mind from wandering and to keep me from getting burned out as I go through something. I take notes in the margins and try to record notes on things I think I'll use again in Zotero. 

[Concordia]

On 9/24/2016 at 1:39 AM, Neist said:

I've been using OCR'ed PDFs...  

Do you use Adobe or some other add-on for the OCR function?  (This is totally new to me, but I have a potential project involving comparing two large documents side-by-side and this could be a useful tool.)

[Neist]

12 hours ago, Concordia said:

Do you use Adobe or some other add-on for the OCR function?  (This is totally new to me, but I have a potential project involving comparing two large documents side-by-side and this could be a useful tool.)

 

I use ABBYY FineReader. It's a little costly, but it's one of the best/easiest to use.

You could use Adobe, but if it's a bad scan, scanned at an angle, or has a complicated layout, you risk the OCR not flowing or generating as it should. 

If you have any additional questions, feel free to ask.  

[Concordia]

Thanks.  I'll see if I can figure out their website.

[MastersHoping]

Great post!

Depends what field you're in, but I'm in political science. A lot of the major texts, particularly at the intro level, can be found on Sparknotes and Wikipedia and other study guides. High school teachers may have shunned Sparknotes, but I found it really helpful to cut down on a lot of extra information to pick out what really matters. 

For journal papers, I'll usually read the title (not as trivial as it sounds!), the abstract, then the intro. From there, I look at the date and try to place it in context with what I know from that era (e.g. if it was published in 1973, I'll know it was published after a certain super important book about Polyarchy but before some other things). The whole time, I'll be trying to make my own guesses about the conclusions and then proceed to read the conclusion. After that, I'll read a few of the middle paragraphs (maybe like a couple paragraphs per page). Oh, and I take notes while reading - I write about 1-2 notes/bullet points for each page/section I read, including page numbers to find it easier later. 

Oh, and for anyone who is coming in to grad school for the first time, don't worry if you can't finish all the readings for a class. Profs know that you can't and won't, just try to get the major themes, issues, and debates from the readings. 

[aberrant]

kinda late to chime but regarding reading habbits:

-. I'm sort of a mix of reading/highlighting/annotating a paper in PDF (create notes) and printouts (write on the margins). Since my lab has a biweekly journal club meeting, I usually printout papers that are either 1) important that I need to check, read, re-read periodically or 2) I wanted to present in journal club. I usually remember papers that I read thoroughly, but I still keep these papers in corresponding folders (e.g. topic, specific technique, etc.), all in .pdfs with supplementary info.

- I usually skim through less significant papers, specially for journal club meeting. Since the expectation from our PI is to be able to summarized a paper (or more) in 5 minutes followed by Q&A of the entire lab, it sort of forced us to learn how to extract the significance of a paper in 5-15 minutes before a lab meeting. I noticed that the reading time shorten as I found myself becoming more comfortable and familiar with the topic / technique. Many of the papers in my field can be understand rather quickly -- read the abstract, understand the figures, read the discussion (sometimes subheadings are suffice), and go through the supplementary info (many gems, such as elaborated method section and extra data, were presented in SI in my field). Figures are extremely important. If I don't have enough time, I would go straight to figures and try to understand the significance of the paper. The key is to be able to point out the significance of a paper, and learn something (hopefully new) from this paper -- for yourself and for your lab.

- For those who are not good at remembering significance/summary of papers, MS OneNote is definitely an option. Although I only use it for my electronic lab notebook, I think it is user-friendly enough to type up bullet points / jot what you read quickly while keeping it organized.

- For citation, I'm a huge fan of Mendeley, even though my boss is a die-hard EndNote fan. Mostly because of user-friendly, editable citation formats, and like someone mentioned, you can read the paper in .pdf while you are organization your list of citation -- if something needs to be check / confirmed quickly while you are writing/editing your paper (I know I don't wanna look through my own folder / printouts of +100 papers).

[MarineBluePsy]

Great suggestions in this thread.  What has been working so far for me is to first organize my reading in order of workload.  Reading attached to homework assignments/papers/presentations/my research is most important, reading for discussion only is next, and supplemental reading that has no assignment or discussion requirement is last.  Second I print all of the articles (double sided, eco saver ink mode, skip the reference pages) and carry a few with me when I have to go to campus.  Scheduled and unscheduled gaps in time are used to read and take notes in the margins.  That way when I get to the assignment/presentation/research my notes are already a summary of key points, I just need to type them up and fill in whatever other information is needed.  Third, in addition to the on the go reading time I schedule time to read any leftover articles or do any textbook reading.  So far this is once or twice a week where I read as much as I can for two hours and that's it.  I love to read, but I find it more productive for me to do it in shorter blocks.  Fourth, readings for discussion only I read the abstract and results and take a note or two.  If it seems like something I might be able to use for a class assignment or my research I'll read the methods.  Lastly, I don't do any of the supplemental reading unless it is something I can use for a class assignment or my research.  To store research articles I use Google Drive and RefWorks.

[Butterfly_effect]

I am a die-hard zotero fan. I just haven't found any other referencing software that has 1. a web plugin that's basically a button you hit when you're on pubmed or jstor or whatever to save the PDF and reference information from the page automatically. I never enter it in myself, though maybe 1/50 times I'll have to fix capitalization or something. 2. an interface that links the PDF files with their reference information so can just double click a reference in zotero and have the annotated PDF pop up in preview and 3. a good word plugin where I can just hit a button to add a citation field, then a search bar pops up and I can type in part of a title or author and select from my library the right paper. Zotero than automatically generates bibliographies in a style you select (e.g. Nature, Cell, Chicago). Does Mendley or Endnote do this? It's so convenient. 

I am also trying to reference and make notes on hard copies as I think I remember papers better this way, though it's harder because I can't search my notes unless I take the extra time to enter them into my annotated bibliography in Evernote. 

[TakeruK]

12 minutes ago, Butterfly_effect said:

I am a die-hard zotero fan. I just haven't found any other referencing software that has 1. a web plugin that's basically a button you hit when you're on pubmed or jstor or whatever to save the PDF and reference information from the page automatically. I never enter it in myself, though maybe 1/50 times I'll have to fix capitalization or something. 2. an interface that links the PDF files with their reference information so can just double click a reference in zotero and have the annotated PDF pop up in preview and 3. a good word plugin where I can just hit a button to add a citation field, then a search bar pops up and I can type in part of a title or author and select from my library the right paper. Zotero than automatically generates bibliographies in a style you select (e.g. Nature, Cell, Chicago). Does Mendley or Endnote do this? It's so convenient. 

Yes, Mendeley does everything you describe  Many paper databases in my field has a "add to mendeley" button but Mendeley also has a Bookmarklet, basically I click a button on my browser and it adds the article I'm looking at into my Mendeley Library, so it works even when the website I'm using doesn't have a Mendeley button. However, my own favourite way of adding papers is to download the PDF into a special folder I've set up. Mendeley knows when a new PDF is added to this folder and automatically adds it to my Library. Usually, the bibliographic information is correct (90% of the time). In the cases where it isn't correct (or not complete), 99% of the time, Mendeley got the DOI right. I then just press a button next to the DOI in Mendeley and it searches the web for this DOI and automatically corrects all the metadata. 

The caveat is that I know a Word plugin exists but I have not used it myself. I use Mendeley to generate .bib files (for BibTeX) and then I generate bibliographies using BibTeX. This is because my field generally uses LaTeX instead of Word for our articles etc. Ultimately, the effect is the same---I never type out bibliographical entries by hand and I can generate bibliographies in whatever style I need.

[Butterfly_effect]

Just now, TakeruK said:

Mendeley knows when a new PDF is added to this folder and automatically adds it to my Library. Usually, the bibliographic information is correct (90% of the time)

Ooh, this is a nice feature. 

1 minute ago, TakeruK said:

I use Mendeley to generate .bib files (for BibTeX) and then I generate bibliographies using BibTeX.

But this way, you'd have to select which publications you used an in what order? Or can you feed a document into Mendeley and have it generate the bibliography? One of the things that is nice about a dedicated word plugin is that I don't have to keep track of anything. I just cite as I write and hit a button at the end to generate a bibliography. It's particularly nice for long papers. 

[TakeruK]

Just now, Butterfly_effect said:

But this way, you'd have to select which publications you used an in what order? Or can you feed a document into Mendeley and have it generate the bibliography? One of the things that is nice about a dedicated word plugin is that I don't have to keep track of anything. I just cite as I write and hit a button at the end to generate a bibliography. It's particularly nice for long papers. 

I write my papers in Latex and use Bibtex to generate bibliographies. It might be more clear if I just give an example .

First, in my Mendeley Library, let's say I have papers from Smith et al. 2002, Jones et al. 2014, and Wong et al. 2011. One thing I need to do is to assign a "ID" to each of these papers. I use the format "Smith2002", "Jones2014" and "Wong2011" etc. (but others may use different conventions). I do this as soon as I import every paper into my library.

Second, I start writing my paper. At the beginning, I tell Latex which BibTeX "style" to use. This includes both things like how the citation is formatted in-text, how the end bibliography looks and whether I want things to be listed in the the order I cite them or in alphabetical order etc.

Third, I write my paper. I might write sentences like "Apple pies are shown to be more effective with ice cream than cherry pies \cite{Wong2011}. However, more people prefer cherry pies with ice cream than apple pies with cheddar \cite{Jones2014}. Although this discussion on pie is interesting, \citet{Smith2002} showed that cakes are vastly superior to pies, whether they are served with ice cream or cheddar." In the actual text, the \cite{} commands are replaced with citations in whatever style I defined above.

Finally, at the end, there is a command to generate the bibliography and it is generated in the same style I indicated above. If I chose a numerical ordering, adding references will update all the numbers too. I can also change the style back in step 2 and it will automatically change all of the in-text and end-of-text styles!

So, I think both methods are functionally equivalent

[aberrant]

On 11/4/2016 at 11:53 AM, Butterfly_effect said:

But this way, you'd have to select which publications you used an in what order? Or can you feed a document into Mendeley and have it generate the bibliography? One of the things that is nice about a dedicated word plugin is that I don't have to keep track of anything. I just cite as I write and hit a button at the end to generate a bibliography. It's particularly nice for long papers. 

 

On 11/4/2016 at 11:49 AM, TakeruK said:

The caveat is that I know a Word plugin exists but I have not used it myself. I use Mendeley to generate .bib files (for BibTeX) and then I generate bibliographies using BibTeX. This is because my field generally uses LaTeX instead of Word for our articles etc. Ultimately, the effect is the same---I never type out bibliographical entries by hand and I can generate bibliographies in whatever style I need.

Just wanted to point out that the word plugin for Mendeley can do what @Butterfly_effect was looking for. That being said, Mendeley can also automatically arrange/adjust the order of citation in bibliography (that is journal / citation format-dependent. You can change your format from 1 journal [e.g. Nature] to another [e.g. PNAS] in a few clicks.)

[adventurek]

On 2016/11/5 at 0:31 AM, Butterfly_effect said:

3. a good word plugin where I can just hit a button to add a citation field, then a search bar pops up and I can type in part of a title or author and select from my library the right paper. Zotero than automatically generates bibliographies in a style you select (e.g. Nature, Cell, Chicago). Does Mendley or Endnote do this? It's so convenient. 

EndNote also does automatic bibliography generation with a large variety of styles.