Google for research: If it’s wrong, why does it feel so right?

In the uni library world we encourage students to use the library’s discovery layers and database interfaces to search for information. We tell first-years over and over again not to use Google. Is this the right thing to do?

Here’s an information retrieval story from today when I was looking for journal articles on a certain topic:

  1. Used library discovery layer. Didn’t like my results.
  2. Used most recommended database #1, which is small but specialised. Simple 2-term search connected by AND. Some success, but not exactly a jackpot of relevant articles, but a few that were of interest.
  3. Added some synomyms to my search strategy to broaden. Results were the same.
  4. Used most recommended database #2. Large, but multidisciplinary. Had to add more terms to refine the search and experiment with my keywords a bit more. Not much success.
  5. Went back to discovery layer and fiddled with my keywords a bit more. Still not satisfied.
  6. Went to Google. Put in my 2 keywords, no syntax or synonyms or anything. Based on the nature of my search terms, suggestions from Google scholar appeared at the top of the results list. It suggested two articles that were basically my ~dream articles~ in terms of relevance, and highly cited.
  7. Clicked on them. Paywall.
  8. Copied and pasted the article titles into library search. There they were, in databases I hadn’t looked at in steps 3 or 5. Clicked through to full text, and downloaded the pdfs in all their glory.

Now, Google was by far the most helpful tool in terms of discovery (step 6). It was very simple. I didn’t even intentionally go to Google Scholar, just plain old Google. The Scholar results were presented right in my face, there was no effort on my part in doing this. But in terms of access, Google let me down (step 7). I happened to know there was a good chance that the library would have access, so I went looking there, once I had the article titles.

The databases and library search tool were not as good at discovery. It was trickier and more frustrating to find stuff,  I had to use some advanced search strategies, and common techniques like broadening my terms didn’t always work (steps 2-5). The few relevant results that I did get were not as good as what I later found on Google. Of course this is not always my experience.  It depends on the amount of literature available on your topic, how well your search terms match the vocabulary of the databases and the literature, and endless other factors. My topic happened to be a little bit niche on this occasion, which is the kind of situation when I think Google provides a better search experience. The real value in the library search and library-subscribed databases was the access itself (step 8).

I know this is just one anecdote. But c’mon, Google can be a life saver sometimes. Let’s not demonise it. Maybe next time a student is stuck in a rut not finding relevant information in databases, rather than complicate the search strategy, just use the Google workaround!

Advertisements

#BlogJune 20 – World Refugee Day

The 20th June is World Refugee Day.

I don’t intend to get political too often on this blog (though it is tempting during election season), but Australia’s treatment of asylum seekers is a consistent source of frustration and heartbreak for me.

A group that I have a huge amount of respect for is Love Makes A Way, who coordinate non-violent protest rooted in the Christian faith. One of their main activities is organising sit-ins in MPs offices, where they pray, sing hymns, and demand to know when children will be released from detention. Civil disobedience is sometimes an appropriate tactic to draw attention to injustice. For those involved in Love Makes A Way It has resulted in arrests in some cases. They have stories about police feeling terrible for arresting pastors, priests, nuns, and the like. But it just goes to show how seriously inhumane our government’s policies are that it would motivate such people to break the law.

you have to understand,
that no one puts their children in a boat
unless the water is safer than the land

-Warsan Shire, “Home”

Let them stay.

#BlogJune 16 – Hotdesking

 

desk-1283688_1920.jpg

Disclaimer: stock photo – not my desk. A library would never use Macs

This week I have inherited my own desk at work.

I work weird hours and I spend more than half of my working hours out on the service desk, so I was previously hotdesking with a few other part time staff. I didn’t mind it, I’m not particularly possessive of having my own space since all I really need to do my job is a personal network drive to store my files, and an internet connection. According to my experience, here are the pro’s and cons of hotdesking compared to… erm, regular desking.

PROS

  • It is a space saver. The whole hotdesking thing was actually initiated when a few staff moved to another campus, meaning there were just enough desks for two offices to be consolidated into one, if a handful of part time staff shared a few desks. The empty office was actually converted to make more student space in the library.
  • Better team relationships. The two old offices were divided into the “Librarians” office and the “Loans team” (i.e. library technician & assistants) office. We are lucky to have a really cohesive staff team, and having everyone together in the same office has fostered that even more. Us librarians used to be in an office upstairs, the “ivory tower”, as I called it, which meant we were not that approachable.
  • Less clutter. You’re less inclined to leave piles of paper around when a  space isn’t “yours”. I’m a self-declared paper-hater, and turn down offers of booklets and handouts and bits and pieces which I’m guaranteed to never look at again. Having limited space to store unnecessary papers is a plus, in my opinion, because it discourages it from accumulating.

CONS

  • It’s hard for people to get in contact by phone since they don’t know which desk you’re at. We have a shared office space so colleagues can easily approach each other in person, but if a phone call comes through the main desk, the staff member there does not know what number to forward it to because they don’t know which station I’m at. I’m sure that workplaces that have embraced “full hotdesk” are mobile based.
  • Using a handful of different computers meant setting up my internet bookmarks for each of them. Even though I saved my bookmarks as an HTML file and imported them to each computer I would use, I would end up adding and subtracting bookmarks and they ended up being all different on different computers. I’m sure there are web-based bookmarking apps I could have used to get around this, but I didn’t bother. I’m not a huge bookmarker, I just like having a  little toolbar of my most used sites to access quickly. Again, I’m sure “full hotdesk” workers use laptops.
  • You’re not able to personalise your workspace. As I mentioned, I don’t mind, but some people find it hard to adjust to different environments all the time, and would rather have their workspace set up according to how they like it and how they work efficiently. Also, it is nice to have pictures, decorations, trophies, etc. to customise your workspace to make it more personal. My first thought when I got set up at my new desk was that I’ll need to bring in some knick-knacks to put around the place.

OTHER OBSERVATIONS

  • Even though there was a handful of desks to chose from, most of us would end up having a de facto “main” desk that we gravitated towards. Mine was the smallest, which made people feel sorry for me, but I didn’t care because I had all the space I needed. Much more than my study at home.
  • Staff members would express confusion and surprise when I changed location day to day. Sometimes they’d be looking for me, and they’d look at my “usual” spot and not see me, so they’d conclude that I wasn’t around. They’d then get spooked when I suddenly appeared sitting somewhere else. Now they’ll know where to find me!

23 Research Data Things – Thing 8: Citation Metrics for Data

Alternative metrics or altmetrics count the number of views, number of downloads, social media “likes” and recommendations associated with a dataset. Because of their immediacy, altmetrics can be an early indicator of the impact or reach of a dataset; long before formal citation metrics can be assessed.”

Do you think altmetrics for data have value in academic settings?  Why?

Altmetrics measure popularity, which is not necessarily the same as “impact” for research. Sometimes a research article will gain a lot of attention outside academic circles. A mainstream newspaper or website may write about it if it’s something likely to capture the interest of the general public. But this is more likely to happen to articles that are controversial, so it doesn’t necessarily measure their impact or importance in the research literature.

However, it’s still a valuable thing to measure, especially for data. Altmetrics are a good match for data sharing in open access repositories and online because they measure web-based engagement. In the current research climate there are very important interactions taking place on social media, and these shouldn’t be ignored or dismissed as something flippant.

23 Research Data Things – Thing 7: Data citation for access & attribution

Data citation is a relatively new concept in the scholarly landscape and as yet, is not routinely done by researchers, or expected by most journals. What could be done to encourage routine citation of research data and software associated with research outputs?

In my uni work I need to reference materials in APA style. I have not yet referenced a data set. The APA Style blog has an entry on how to cite a data set. It’s given the “miscellaneous object” treatment, in that you need to specify its format, [Data Set], in a way you don’t need to for journal articles, books, or even webpages. To me this signifies that it’s still a “rare” type of citation.

In terms of building a culture of data citation, I think giving a dataset a DOI goes a long way to “legitimise” the document in the minds of readers and make people feel more inclined to cite it.

It may feel shallow, but I’m convinced that presenting data in an attractive way makes it more likely to be cited. The visualisations created with infogram are pretty sexy. Data visualisations are just easier to read and make the material more engaging. Also having a nice layout makes them look “pro”, while looking at an excel spreadsheet may (even subconsciously) give a reader a sense that the work is incomplete.