Preliminary review of survey responses

[arwenhutt]

Ideas on approaching analysis

[rmjaffe]

I can help with this too if it proves to be more than a one person job.

[kbalster]

Thanks!

I'll try to take a look at the responses next week, and if we start getting in some more, then I'll get in touch. I figure we can work out a best approach for analysis over email.

[rmjaffe]

Sounds good. I'm wondering -- should we include the use cases from our Github issues among the survey results?

On Fri, Dec 15, 2017 at 1:48 PM, kbalster notifications@github.com wrote:

Thanks!

I'll try to take a look at the responses next week, and if we start getting in some more, then I'll get in touch. I figure we can work out a best approach for analysis over email.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ucsdlib/uc-linked-data/issues/17#issuecomment-352120299, or mute the thread https://github.com/notifications/unsubscribe-auth/AehVTxjIRFgQ3L302j76NOLl5dzNr3zvks5tAulDgaJpZM4RD65p .

-- Rachel Jaffe Metadata Librarian Metadata Services, University Library University of California, Santa Cruz 1156 High Street Santa Cruz, CA 95064

jaffer@ucsc.edu (831) 502-7291

http://library.ucsc.edu/

[arwenhutt]

@rmjaffe @kbalster I think we'll want some separate analysis or summary for the "what's happening in the system" part of our charge, but they would also be a good source for mining additional use cases to add to, or for showing interest/support for, the ones we already have.

[katstine]

I'm happy to jump in here (albeit a bit late) with analysis help if you need it!

[rmjaffe]

The more the merrier, methinks!

On Fri, Dec 22, 2017 at 11:46 AM, katstine notifications@github.com wrote:

I'm happy to jump in here (albeit a bit late) with analysis help if you need it!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ucsdlib/uc-linked-data/issues/17#issuecomment-353663845, or mute the thread https://github.com/notifications/unsubscribe-auth/AehVT4FHFX470E_1OQp3epNX1G0MiIN4ks5tDAcDgaJpZM4RD65p .

-- Rachel Jaffe Metadata Librarian Metadata Services, University Library University of California, Santa Cruz 1156 High Street Santa Cruz, CA 95064

jaffer@ucsc.edu (831) 502-7291

http://library.ucsc.edu/

[kbalster]

Hi @rmjaffe and @katstine, it looks like we probably have all the survey results we can expect, so I think we can get to work on the analysis. There are 17 total responses, so it should not be too difficult. I am taking a first pass through the response spreadsheet https://docs.google.com/spreadsheets/d/13btSziox771xlxh4Zbt6G__V1ipqq28fyTOzfltyfM8/edit#gid=1518377035(url) and added a second tab for analysis if you'd like to see what I have so far. Any suggestions you have for grouping together results would be much appreciated, and feel free to make any edits to the analysis tab. I suspect we could have something put together for the meeting this Friday, and since these results will likely be used to support our first webinar, keep an eye out for any details, quotes, etc. that really stand out.

Thanks,

Kevin

[katstine]

hi @kbalster! thanks so much for setting up the analysis sheet - it's great! I added three columns at the end: "Needs", "Goals", and "map to Use Case". In Needs, I'm trying to track where mentions have been made of systems or tools that might be useful in supporting campus' work (beyond just what they've noted in response to the "challenges" question). I'm also working in there to try to distill/code the goals represented by each response to a standard set of terms that we can use for further analysis. Haven't done anything yet with the Use Case column, but thought it'd be helpful for us to have a placeholder for future analysis/synthesis work.

[chrissyrissmeyer]

@kbalster @katstine @rmjaffe: My apologies, but I only submitted my response to the form today (we have, er, had a lot going on in SB the past month). Hoping one of you saw this.

[chrissyrissmeyer]

@kbalster @katstine @rmjaffe: just saw another response came in from UCSB this afternoon (I reminded some folks this week) -- sorry again for the late responses!

[rmjaffe]

@kbalster @katstine I'm wondering if we should hold off on any further analysis until we've contacted and (hopefully) heard back from the people and projects we targeted in today's meeting?

[katstine]

@rmjaffe @kbalster in that case, should we impose a new deadline in that case? also, I'm happy to meet up by Zoom with you all to discuss findings, take a stab at what/how to present, etc.

[rmjaffe]

@katstine, I definitely think a new deadline is a good idea. How long should we give and how much time do we need to give ourselves to have our data and findings in order before UC-DLF? A Zoom call is another great idea!!

[chrissyrissmeyer]

Just emailed Charlie @ UCSF. I'll update here once I hear back from him.

1/19 - Charlie submitted a response and said that to the best of his knowledge there is no linked data work currently happening at UCSF Library. He did forward my emails others doing more specific work with the library’s various Data Science initiatives just in case, so perhaps we will get a response from them as well.

1/19 update - Also received confirmation from Susan Boone, Technical Services Operations Manager at UCSF Library that they have no firm plans at this point.

[ngeraci]

Just emailed Laura (UCI) & will likewise update.

1/16 - Laura just submitted a response

[rmjaffe]

@katstine @kbalster Just wanted to check in -- when are you guys available for a Zoom chat?

[katstine]

yes! this Friday, anytime 9-11am or 12-2pm. then next week Tuesday 1/23 11am-2pm or 3-5pm. the remainder of the week is pretty crazy.

messaged Brian Tingle (CDL) and Emily Lin (UCM), asking if they could reply to the survey by next week.

[rmjaffe]

Great! How does next Tuesday 3-4 sound? @kbalster?

[kbalster]

Ugh, sorry for falling off the face of the earth this past week @rmjaffe @katstine @chrissyrissmeyer. My schedule is pretty open next week to set up a Zoom call (any time after 11:00 Monday-Friday).

[rmjaffe]

Totally OK @kbalster , it happens :) But I do think we should try to arrange for something next week.

I am free after 2:00 Tues-Fri next week.

[katstine]

draft slide deck with analysis results: https://docs.google.com/presentation/d/1tPh1ytWWfSyxsm0xYrNZ7uQxL62UHhN5yBDoa-kAhoQ/edit?usp=sharing

[rmjaffe]

Hi All,

I finished looking at the responses for collaborators, functional areas, and technology. I wasn't sure how to best integrate my work into Kevin's worksheet, so I added a new worksheet, RJ Analysis Collaborators, Areas, Tech.

I experimented with pulling apart the data, specifically the collaborator information, in a couple different ways (possibly useful or interesting, possible not). I put my "findings" in separate document: https://docs.google.com/document/d/1DPMtKI_jzyZihZMb-QTyPi3xyZjvY0b228QxB2Z4Mvo/edit?usp=sharing

I didn't add them to the draft slide deck because I'm not sure how this information would be best presented: visually, as quick bullet points, etc. I welcome your feedback @kbalster @katstine @chrissyrissmeyer

[kbalster]

Hi @rmjaffe,

This looks great! It is certainly easier to interpret and work with than the worksheet I created. I will add columns for the remaining categories.

Kevin

[kbalster]

Hi all, just wanted to provide an update. I got the raw metadata column in, but I've been out of the office the past 4 days, so I haven't been able to complete the metadata analysis column or add the other free text columns. I can finish up the metadata column once I get back on Monday, but feel free to take a first pass at the free text columns.

[rmjaffe]

@kbalster Just a heads up: I started fiddling with the Which metadata standards/schema? results.

[katstine]

@rmjaffe - thanks so much for what you've extracted and analyzed from the spreadsheet into the Google doc. Seems like much of this will be really interesting to share at the BoF, with the quantitative data perhaps going into pie charts? And the points of interest that you've pulled together perhaps serving as opportunities to encourage discussion, more feedback from attendees.

[kbalster]

@rmjaffe and @katstine I took a first pass at the goals, challenges, conclusions, and successes columns, including analyses with a few controlled phrases. Fire away!

[arwenhutt]

@kbalster @rmjaffe @katstine Thank you all for your work, all of these views of the survey are really useful! I've been looking at them and playing with some different ideas for pulling out some numbers/visuals to represent some of the insights you all have already made, and some things I find interesting too. I'm not working in the google sheet though because it is so slow, but I'll add anything I think might be useful to the ppt.

I feel like we have a lot of good content for the presentation (more probably than 15 min) and am feeling pretty good about getting it pulled together in the next couple of weeks.

[cstahmer]

2018_02_21_uc_linked_data_survey_summary.zip

Here are some plots that I drew that summarize some of the responses.

[arwenhutt]

Thanks @cstahmer these are great! Are these based on text analysis of the original responses?

[cstahmer]

There isn't enough data to do any really fancy NLP or anytying. We broke each question down into a Term Document Matrix where we could identify counts for meaningful words by answer and then referenced these back to institutional metadata for the responses. I think they give a fair idea of what's going on for the topics we covered. We also tried doing a sentiment analysis on all the longer answers to see if we could glean anything about whether respondents were relating positively or negatively to their answers, but this was inconclusive (and overkill, since there were so few responses that you could figure this out just by reading anyway.)

arwenhutt mailto:notifications@github.com February 22, 2018 at 12:51 PM

Thanks @cstahmer https://github.com/cstahmer these are very great! Are these based on text analysis of the original responses?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ucsdlib/uc-linked-data/issues/17#issuecomment-367817803, or mute the thread https://github.com/notifications/unsubscribe-auth/ACEe9RhYfBCd9XFXeZXxBGT9aIE2D3Rfks5tXdNUgaJpZM4RD65p.

arwenhutt mailto:notifications@github.com December 15, 2017 at 11:46 AM

Ideas on approaching analysis

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ucsdlib/uc-linked-data/issues/17, or mute the thread https://github.com/notifications/unsubscribe-auth/ACEe9fPPyNF1lLVZJj1BPFRx2iVRfCkBks5tAsyOgaJpZM4RD65p.

-- Carl G. Stahmer, PhD Director of Data and Digital Scholarship UC Davis Libraries Associate Director for Humanities UC Davis Data Science Initiative Associate Director English Broadside Ballad Archive www.ds.library.ucdavis.edu www.carlstahmer.com @cstahmer

[katstine]

thanks, @kbalster and @cstahmer! it's interesting comparing the linked words for the goals/challenges/conclusions questions against the manual coding and analysis we've done.

[arwenhutt]

Slide deck https://docs.google.com/presentation/d/1qxHUz7vFkolVc9YHLARxosgqz8ibja3KMl9IL0CsTio/edit?usp=sharing Spreadsheet with massaged data used for charts https://docs.google.com/spreadsheets/d/1EEvHzCT1hbatYheAp5vj1uhZpBz-koGRkGOIxcQ9CZM/edit?usp=sharing