I gave the second 3-hour "course" to the Haiti Met Service on 18th September 2018. This was on simple climatic analyses. It worked much better than the first. This is the first time I have tried to run such a course over skype/zoom and I suggest it is possible formula more generally.
a) The data were "clean". They were daily rainfall for 3 stations. The first course (31 August) had to concentrate on tools for cleaning the data in R-Instat. This was difficult and not very satisfying for the participants. (This was partly the task, and partly because of the number of steps involved in the current R-Instat.)
b) We used zoom this time, rather than skype. It worked very well. There were 3 "participants" in Reading, Florida and Haiti. We didn't use video and I don't know whether we could have. Though we shared the screens so it wouldn't have made that much difference.
c) We shared the screen. In the first session we shared my screen, and this was easy for me. This time we shared one of their screens and I think this worked better. They sometimes stopped after a dialogue to repeat locally on the furst machines and that was fine too.
d) The sequence in the climatic menu works well.
i) We imported the data from Excel (File > Open)
ii) Then used the climatic tidy menu to put the data into shape, to check, then to correct.
iii) Then the climatic dates menu to infill missing years, then add columns for year, month, doy, etc.
iv) Then the climatic check data menu for the inventory and boxplots.
v) Then the Climatic Prepare menu to get annual summaries (rainfall totals).
vi) Then the Graphics Line plots to show a PICSA-style graph. (Soon be able to use the Climatic PICSA menu.)
vii) Then the Describe > Calculate menu to get a column for the rain days
viii) Then summarised the rain days and changed the graph to show this.
e) The only difficulty I found was the explanations of where to go, both in the menus and within a dialogue. This was partly because I was doing it in my limited French, but that was not the only reason. I suggest it could be helped by us developing "standard terminology" and then perhaps having a few screen-shots of menus and a few dialogues to explain the terms.
f) As with an earlier session over zoom (by Stats4SD and not climatic) there was a second resource person. This was Caroline Staub, and she was in Florida. She set up the zoom connection, helped to sort out the issue in Haiti when the participants had difficulty sharing their screen, and also shared the discussion and questions. It is very useful having two people in this way.
However, overall I thought it went well. It is (for me) a new way of giving a climatic course. There will be a third session and then they will give a local course within Haiti.
I gave the second 3-hour "course" to the Haiti Met Service on 18th September 2018. This was on simple climatic analyses. It worked much better than the first. This is the first time I have tried to run such a course over skype/zoom and I suggest it is possible formula more generally.
a) The data were "clean". They were daily rainfall for 3 stations. The first course (31 August) had to concentrate on tools for cleaning the data in R-Instat. This was difficult and not very satisfying for the participants. (This was partly the task, and partly because of the number of steps involved in the current R-Instat.) b) We used zoom this time, rather than skype. It worked very well. There were 3 "participants" in Reading, Florida and Haiti. We didn't use video and I don't know whether we could have. Though we shared the screens so it wouldn't have made that much difference. c) We shared the screen. In the first session we shared my screen, and this was easy for me. This time we shared one of their screens and I think this worked better. They sometimes stopped after a dialogue to repeat locally on the furst machines and that was fine too. d) The sequence in the climatic menu works well.
i) We imported the data from Excel (File > Open) ii) Then used the climatic tidy menu to put the data into shape, to check, then to correct. iii) Then the climatic dates menu to infill missing years, then add columns for year, month, doy, etc. iv) Then the climatic check data menu for the inventory and boxplots.
v) Then the Climatic Prepare menu to get annual summaries (rainfall totals). vi) Then the Graphics Line plots to show a PICSA-style graph. (Soon be able to use the Climatic PICSA menu.) vii) Then the Describe > Calculate menu to get a column for the rain days viii) Then summarised the rain days and changed the graph to show this. e) The only difficulty I found was the explanations of where to go, both in the menus and within a dialogue. This was partly because I was doing it in my limited French, but that was not the only reason. I suggest it could be helped by us developing "standard terminology" and then perhaps having a few screen-shots of menus and a few dialogues to explain the terms. f) As with an earlier session over zoom (by Stats4SD and not climatic) there was a second resource person. This was Caroline Staub, and she was in Florida. She set up the zoom connection, helped to sort out the issue in Haiti when the participants had difficulty sharing their screen, and also shared the discussion and questions. It is very useful having two people in this way.
However, overall I thought it went well. It is (for me) a new way of giving a climatic course. There will be a third session and then they will give a local course within Haiti.