Global literature review - Google Scholar search notes
Notes from Jonah Olsen
As we discussed, it probably makes the most sense to have our keywords decided on from the beginning so that we can be consistent through our searches. The framework I came up with s based on the objectives outlined in the proposal.
In coming up with these keywords, I made sure to test them to find which produced the most (relevant) results on UM libraries. I took for granted that accompanying each keyword(s), “dried fish” (with quotation marks) would be included. However, I also noticed different sources come up when I used “dry fish” instead, so that may be something to discuss (I am not sure if there is a different association with that term). Finally, as we discussed, it is important to have a disciplinary and methodological keyword attached to each article. However, after my test searches, it seemed that attaching that kind of info as a search term did not yield relevant results, so it might be better to hold off on determining those keywords once we are going back through the sources and looking at abstracts. This means, unfortunately, that we won’t have those keywords pre-established, but I don’t see that being a big problem (it will be quite obvious when looking through which disciplines are underrepresented).
As I mentioned in our meeting, I think it makes sense to have keyword theme-clusters, so that there is some uniformity (i.e. each paper will have at least one keyword from each cluster). The clusters and sub-keywords I have identified (which could also be a way of breaking down the review sections) are as follows:
- Discipline (to be further assessed once we have collected our sources)
- We can subdivide this by the type of data/method once this has been determined
- Regional focus (to keep this open, I would suggest we include all the continental, national, regional, and local focuses of the papers here; e.g. if the focus is on Bangkok, include the keywords Asia, Southeast Asia, Thailand, and Bangkok). Include most/all of the following:
- Continent (specifically)
- Country (specifically)
- Region (specifically)
- Locality (specifically)
- Geographical Scale (to indicate how broad the study is). Likely include just one of the following, unless it deals with different levels in depth.
- Global
- Continental
- National
- Regional/Provincial
- Local
- Coastal vs inland
- Chronological Focus (we may want to discuss whether we want to do this by decade, year(s), or just a binary split between historical cases and contemporary cases).
- When it was written (which will be useful for the historiographical element of the review paper)
- Probably best broken down by decade
- Where on the value chain (this will be something else that we can only assess after looking at the abstracts, so we will have to wait to determine the specific keywords we will use to tag).
- Focus on supply vs demand
- Problem being addressed
- Wellbeing (of communities), which could be further subdivided into the following (assuming they are not directly using a wellbeing framework but still address some aspect(s) of it
- Material assets
- Social relations
- Subjective perceptions
- Nutrition (including food security etc.)
- Ecology
- Labour (fisher/worker wellbeing)
- Maybe this should be split between fishers and workers on land
- Wellbeing (of communities), which could be further subdivided into the following (assuming they are not directly using a wellbeing framework but still address some aspect(s) of it
- Optional keyword tag if they take one of the specific theoretical orientations mentioned in the proposal (I see these as separate from the disciplinary focus)
- Diverse economies
- Value theory
- Political ecology
- Critical feminist studies
- Social wellbeing
So, an article that takes an ethnographic approach to working conditions among fisher-driers in coastal Andhra Pradesh might have keywords tagged as follows:
Anthropology, Ethnography, India, Andhra Pradesh, Regional/Provincial, [Place on Value Chain], Coastal, Supply, Labour, (Optional) Social Wellbeing
Here are the keywords/search methods I found to be the most helpful, based on searches on the UM Libraries website. These do not have to be the keywords we use to tag the papers on Zotero, of course, they just help to give a sense about what’s out there for us. As we develop this, I think we will find some words that are better as search terms and others that are better as keyword tags.
| Keyword 1 | Keyword 2 | # of Results |
|---|---|---|
| “Dry fish” | N/A | 2,242 |
| “Dried fish” | “Dry fish” | 247 |
| Dried fish | N/A | >235,000 |
| “Dried fish” | N/A | <10,000 |
| “Dried fish” | Asia | 1,776 |
| “Dried fish Asia” | N/A | 0 |
| “Dried fish” | South Asia | 1,233 |
| “Dried fish” | “South Asia” | 188 |
| “Dried fish” | “Southeast Asia” | 413 |
| “Dried fish” | India | 1,820 |
| “Dried fish” | Myanmar | 115 |
| “Dried fish” | Burma | 205 |
| “Dried fish” | Bangladesh | 399 |
| “Dried fish” | Cambodia | 194 |
| “Dried fish” | Thailand | 637 |
| “Dried fish” | Andhra Pradesh | 49 |
| “Dried fish” | Gujarat | 75 |
| “Dried fish” | Sri Lanka | 260 |
| “Dried fish” | Kerala | 82 |
| “Dried fish” | West Bengal | 212 |
| “Dried fish” | Community | 2,765 |
| “Dried fish” | Local | 4,107 |
| “Dried fish” | Regional | 1,639 |
| “Dried fish” | Region | 3,293 |
| “Dried fish” | Continent | 836 |
| “Dried fish” | Value | 4,531 |
| “Dried fish” | “Value Theory” | 5 |
| “Dried fish” | “Value Chain” | 70 |
| “Dried fish” | “Value Chains” | 33 |
| “Dried fish” | Network | 1,257 |
| “Dried fish” | Economy | 1,928 |
| “Dried fish” | “Political Economy” | 331 |
| “Dried fish” | “Political Ecology” | 39 |
| “Dried fish” | “Diverse economies” | 1 |
| “Dried fish” | Gender | 871 |
| “Dried fish” | Women | 3,415 |
| “Dried fish” | Woman | 2,162 |
| “Dried fish” | Feminist | 140 |
| “Dried fish” | Feminism | 65 |
| “Dried fish” | Class | 2,387 |
| “Dried fish” | Labour | 2,392 |
| “Dried fish” | Technology | 2,326 |
| “Dried fish” | Health | 3,646 |
| “Dried fish” | Food | 6,728 |
| “Dried fish” | “Food security” | 391 |
| “Dried fish” | Trade | 2,959 |
| “Dried fish” | Wellbeing | 106 |
| “Dried fish” | Nutrition | 1,619 |
| “Dried fish” | Ecology | 899 |
| “Dried fish” | Supply | 2,739 |
| “Dried fish supply” | N/A | 1 |
| “Dried fish” | Demand | 2,191 |
| “Dried fish demand” | N/A | 1 |
| “Dried fish” | Survey | 2,221 |
| “Dried fish” | “Fisher wellbeing” | 1 |
Work log from Jonah Olsen, 2018
May 1-5
I began by searching with different keywords attached to “dried fish” and “dry fish” and recording the number of results for each to get a sense of how much material was out there already. This was done on the UMLibraries website. Seeing how many results were generated, I realized that I needed stricter criteria (outlined below).
May 6-19
I have been conducting the searches on Google Scholar using the U of M proxy. I began by collecting all of the sources listed in the bibliography of the proposal (although a number were unavailable, which I have noted).
I have saved articles that are either very clearly focused on dried fish (particularly with some relation to the value chain) regardless of regional focus, and those providing some discussion of dried fish in one of the focus regions. By far the most material is related to Bangladesh, but there is also a significant body of literature on India and Cambodia, followed by Myanmar (although much less on Burma). There is very little literature on Sri Lanka that is not purely scientific. I have found that, often, the best sources are a few (sometimes 10 or more) pages in on Google. Often “value chains” is an effective search term, yielding results from Africa and Asia and Bangladesh in particular. I tried attaching discipline terms such as anthropology and ethnography but that led almost exclusively to archaeological research on prehistoric North America.
Based on our discussion, I have been searching the same keywords (region, value chains, etc.) but with “fermented fish”, “smoked fish”, and “cured fish” as well. These have been yielding some results, although not as many as “Dried fish” and “dry fish”.
I have gone back to all of the previous finds and tagged them with the search terms I used to find them. I have also added bibliographic information to the files that did not have metadata. I am now saving all of the search terms I used along the way to the files after each search.
I have also started to revise the keywords (for tagging in Zotero) so that we can have specific tags to add based on the literature I have found, which will be helpful in sorting the studies.
May 20-31
I often found the same sources using a wide range of search terms. I continued to save them, but with the new search terms. I then merged them so that the library did not become flooded, while still showing that the source came up with each set of search terms.
Generally, “Dried Fish” (followed by the secondary search term) produced the most results, although this is in part due to the fact that I always used it first. Despite this, “Dry Fish” frequently led me to results that I would have otherwise missed, suggesting that the two terms may not be used interchangeably in the literature. Some search terms pointed to recurring themes, such as when the terms ‘“Dry Fish” Cambodia’ yielded a high number of results regarding gender in the fisheries. In every case, “Fermented Fish”, Smoked Fish”, and “Cured Fish” led me to far fewer sources, although the ratio varied significantly by region.
Many sources that I have found are in the gray literature, requiring that I add the bibliographic information manually. Unfortunately, not all relevant information has been available in each case, so some citations are currently incomplete.
