Intelligence Gathering

Today

You walk out with a topic, a set of search terms that actually work, and the start of a search log. That log is most of the Librarian Visit Report.

No librarian is coming to us. You book your own appointment at Lovejoy and go to them, and that meeting is the assignment. Today is how you arrive there with something worth their time.

Have a topic in mind. Rough is fine.

Thirty sources in an hour

• You find one article that looks right. It cites twelve others.
• You chase three of those. Each cites fifteen more.
• An hour later you have thirty sources open and no idea which one to read first
• The instinct is to retreat: narrow the topic, keep three recent articles, hope
• That instinct is completely reasonable, and following it is what makes the next three weeks harder than they need to be

It is a map, not a reading list

A literature review is not evidence that you read enough articles. It is a map of who has said what, and where the arguments are.

Think of it as walking into a debate that has been running for years. Say your piece cold and you get told it was answered in 2012. Knowing the terrain is how you earn the right to ask your question.

You cannot map terrain from three landmarks, which is why retreating to three articles fails.

What a review does for you

• It shows where your work sits. You are not the first person to wonder this.
• It finds the gap, which is your answer to “so what?”
• It stops you rebuilding something that already exists. Your original idea may be somebody’s dissertation from five years ago.
• It hands you methods: measures that have been validated, sampling strategies, and other people’s documented limitations (Krippendorff, 2018; Neuendorf, 2017)
• It sharpens the question from “something about Twitch chat” into something you can actually answer

Phase one: search broadly

• Start wide. You are answering basic questions, not finding your final sources.
• What words do scholars use for this? Who keeps coming up? Which journals publish it?
• Google Scholar first. Less thorough than the real databases, and fast and free.
• Wikipedia is not citable, and its reference list is still a doorway
• Collect anything promising in Zotero and tag freely. Do not tidy yet.

Phase two: search systematically

Now you want coverage rather than speed, which means the real databases:

• Communication & Mass Media Complete, the main one for our field
• PsycINFO if your question touches emotion, persuasion, or cognition
• Web of Science and Scopus for tracking citations across disciplines

These do not work like Google. They expect Boolean operators, which is the next slide.

Building a search string

AND narrows, so both terms must appear. OR widens. NOT excludes. An asterisk is a wildcard: stream* catches stream, streams, streaming, and streamer.

(livestream* OR "live streaming" OR Twitch) AND
(motivation* OR gratification* OR engagement) AND
(viewer* OR audience* OR chat)

Three concept blocks joined with AND, each holding its own synonyms. Build yours the same way and take it with you: a librarian can fix a string like this in seconds.

Keep a search log

Plain text, in your repository. For every search you run, write down:

• The date and which database
• The exact string, copied and pasted, never remembered
• The limiters you set and how many results came back
• How many you kept

You should see it getting boring to write, which is the point. It lets you defend the strategy later, and it shows you fast which searches are not working.

Phase three: follow the citations

• Pick a few keystone articles, the ones that clearly sit at the center
• Backward chaining: read what they cite, which takes you to the foundations
• Forward chaining: use Google Scholar’s “Cited by” to see who has cited them since, which shows you how the argument moved
• It snowballs, but on purpose, following the network out from what matters most rather than wandering

Knowing when to stop

Saturation is the point where more searching stops changing your understanding. You will recognize it three ways:

• Three database searches in a row turn up nothing new that is relevant
• New articles keep citing the same eight or ten sources you have already read
• You can guess what an article says from its title and abstract

It does not mean you read everything. It means stop searching and start synthesizing.

Book your librarian appointment

Do this today, not Thursday

• Go to Lovejoy Library and ask for the Mass Communications subject librarian by name. That is what they are there for.
• Book it this week. The report is due Friday and their calendar is not yours.
• Bring your topic, your terms, and your search log. A consultation with nothing in hand wastes the half hour and gives you nothing to write up.
• This is the part a chapter cannot give you: the interface, the limiters, and the tricks that only come from using these every day.

What to ask while you are there

• Which database is actually right for my question?
• How do I get the full text when the library does not subscribe?
• How do I set up a saved search and an alert for new results?
• What is the fastest way to push a batch of results into Zotero?
• Interlibrary loan is free and much faster than you expect. Ask how to use it.

The rest of today: search, in here, now

• Open a database and run your first real string. I am walking the room.
• Start the search log on your first search, not once you have something good
• Before you leave: a topic, one string that works, and three sources in Zotero
• Bring me the search that returns 40,000 hits, or the one that returns none. Both are normal and both are fixable in about a minute.

Last part of today: commit your work

You have a topic, a working string, and the start of a search log. None of it is anywhere but your laptop.

Download this into the top folder of your workspace, beside README.md:

aura-lab.siue.edu/v2v-hub/starters/01-commit-your-work.qmd

Getting that file onto GitHub is the check. It is not in the template, so if I can see it in your repository, you put it there yourself.

Save, stage, commit, push

1. Save it, Ctrl+S. The dot in the tab turns back into an X.
2. Source Control in the left bar, or Ctrl+Shift+G. Your file is listed with a U.
3. Stage it: hover over the name, click the +. It moves to Staged Changes.
4. Type a message a stranger could read, then click Commit. Never “update”.
5. Click Sync Changes. The first time from a folder it says Publish Branch instead.

You should see the file on your repository page at github.com when you refresh it. A commit stays on your laptop; the push is what sends it. You need both, every time.

Before next time

• The Librarian Visit Report is due Friday, September 18, 50 points. You cannot write it without the meeting, so book the meeting today.
• It wants your search log: databases, strings, results, and what you kept
• Install Zotero and connect it to your browser if you have not yet
• Read the rest of Chapter 4 on the literature map before Tuesday
• Tuesday we turn a pile of sources into a map, and a map into an argument