describe the sample of children in the dataset

[ernestguevarra]

• [x] How many children are in the overall sample?
• [x] How mnay children are males and how many are females?
• [x] How many children are from Urban Montserrado and how many are from Grand Bassa?
• [x] How many children are males from Urban Montserrado and males from Grand Bassa?
• [x] How many children are females from Urban Montserrado and females from Grand Bassa?
• [ ] What is the distribution of ages of children in the overall sample?
• [x] What is the distribution of ages of children by the sex of the child?
• [x] What is the distribution of ages of children by location?
• [ ] What is the distribution of ages of children by sex and by location?

[ernestguevarra]

@OxfordIHTM/eevee

[rahmanianabila]

I have opened the nut data and these are my questions

1. What are 1 and 2 assigned for Sex, muac, and edema?
2. Why are there NA in some of the values?
3. Why are all of them from Urban Montserrado?
4. Are the age unit in months?

[rahmanianabila]

Please cancel question no.3. Thank you

[rahmanianabila]

sex: 1-> male 2->female age in months muac in cm oedema: 1 -> nutritional oedema detected 2-> not detected 99-> no value muac & oedema screen (have your child been measured like this?) 1-> yes 2-> no 99-> don’t know/no response cov-status (is your child in a program that receives peanut butter?) 1-> yes 2-> no 99-> don’t know/no response

[ernestguevarra]

sex: 1-> male 2->female age in months muac in cm oedema: 1 -> nutritional oedema detected 2-> not detected 99-> no value muac & oedema screen (have your child been measured like this?) 1-> yes 2-> no 99-> don’t know/no response cov-status (is your child in a program that receives peanut butter?) 1-> yes 2-> no 99-> don’t know/no response

For muac_screen and oedema_screen, the question asked to the mother is "Has your child been measured liked this (MUAC or oedema) in the past month.

[ernestguevarra]

@OxfordIHTM/eevee, some notes from the questions you and your classmates have asked:

• Other than describing the data by location, sex, please describe the data also by survey round (baseline and endline)
• Watch out for NA values. Some rows of data may not necessarily represent a child that has actually been surveyed. So, discuss as a team how you want to handle these instances (whether to report it in the summaries as children who were visited but not surveyed / refused to be surveyed)

[ernestguevarra]

Team @OxfordIHTM/eevee, I have looked at the branches from each of your team members. I am seeing good things. And I sense as well that you guys were having good discussions around the table during today's session. Keep that up.

I think your problem set is very interesting as this is usually the type of analysis that is presented first in the results section of a research. It might seem very different from the other team's problem set but I think yours is the foundation from which a good research paper is built on.

@ashishlwh-giri and @daisympando has tried some initial code to start addressing the questions for the particular problem set you are dealing with. I have a feeling that all of you have been working together to try out the same code and helping each other. I really like that approach so I look forward to seeing the other team members branches (when you can) and look forward to further questions from you.

I think you got the brief spot on - this is a hackathon and not a race - slow and steady and working together is key!

Keep it up!

[Shuma1510]

sum(nut$sex %in% c(1)) sum(nut$sex %in% c(2)) sum(nut$sex %in% c(NA)) A <- sum(nut$sex %in% c(NA)) B<- sum(nut$sex %in% c(2)) C <- sum(nut$sex %in% c(1)) B+C-A

[ernestguevarra]

Team @OxfordIHTM/eevee, I hope you are doing well.

I made comments to a pull request that @daisympando put in 2 weeks ago. I hope you guys saw that. I am assuming that was the groups submission/pull request? The pull request is erroring on the automated checks and I've mentioned in my feedback where the issues are coming from.

For our last session next week, get your R script completed and then start on preparing your section in the hackathon report. In the project/repository, you would have noticed an .Rmd file named coverage_assessment_report.Rmd. When you open that file, you should see a pre-prepared report format for the results of this hackathon and you specific section outlined in the report.

Using what you have learned in our R Markdown session, start adding text and paragraphs for your section of the report to present the results of your analysis. Please use tables to present the counts for the sample and then plots to show the distribution.

I will walk around/move around and help you get started during our last session.

[ernestguevarra]

sum(nut$sex %in% c(1)) sum(nut$sex %in% c(2)) sum(nut$sex %in% c(NA)) A <- sum(nut$sex %in% c(NA)) B<- sum(nut$sex %in% c(2)) C <- sum(nut$sex %in% c(1)) B+C-A

Here are more appropriate ways of writing the code you have above:

sum(nut$sex == 1, na.rm = TRUE)   ## Much better to directly test equality of the 
                                  ## values rather than using %in%

sum(nut$sex == 2, na.rm = TRUE)

sum(is.na(nut$sex))               ## Remember that NA is a special value in R. 
                                  ## To test if something is NA, we use the function is.na()

[ernestguevarra]

Team @OxfordIHTM/eevee, a table that looks like this on the report will be a good summary of your analysis results:

Urban Montserrado

	Baseline	Endline
Male	Number of males in baseline (% of males out of total baseline)	Number of males in endline (% of males out of total endline)
Female	Number of females in baseline (% of females out of total baseline)	Number of females in endline (% of females out of total endline)
Total	Total number baseline	Total number endline

Grand Bassa

	Baseline	Endline
Male	Number of males in baseline (% of males out of total baseline)	Number of males in endline (% of males out of total endline)
Female	Number of females in baseline (% of females out of total baseline)	Number of females in endline (% of females out of total endline)
Total	Total number baseline	Total number endline

[ernestguevarra]

Team @OxfordIHTM/eevee, I checked again today and I didn't see a team or an individual pull request from any of you. All I have seen is a commit and push from @rahmanianabila on Monday.

As I am trying to complete the review of your code to provide feedback. I will use @rahmanianabila branch as your team's contribution for now but I encourage all of you to make a push of your work on your branch so I can provide individual feedback.

It would be ideal if @rahmanianabila can make a proper pull request as that will make it easier for me to provide feedback. If you can do that today that will be great, otherwise, I will just make a merge of your branch to a temporary branch and then I will make notes of my feedback and put it in the issue page so you all can see it.

[ernestguevarra]

Team @OxfordIHTM/eevee, thank you for your work on the hackathon.

You can now view the overall report with your contributions at https://oxford-ihtm.io/ihtm-hackathon-2023/coverage_assessment_report.html.

Given that the team didn't make a formal and complete pull request, I used the branch of @rahmanianabila as she was the latest one from all of you who has made a commit and push to her branch and comparing her branch with the branches from all other team members, she has the one with code/solutions for most of the questions. Without a pull request, I cannot make the typical feedback/review process via GitHub so I am using this issue comment to provide you team feedback. I will provide individual feedback to each of you separately.

Following are my comments and feedback to the team with regard to your efforts and contributions to the IHTM Hackathon 2023:

• I acknowledge the team effort shown by all of you for this hackathon. Based on my observations during the hackathon period specifically during class, your team was the one always huddled together and working with each other and talking with each other. The last day was a bit different as half of your team were not in but you still collaborated and got things through.
• I am impressed with the use of {ggplot2} package for plotting. My guess is that some of you have encountered it in your other classes (modelling class?) and tried to apply it here or that you went out of your way to learn it. See the edits I made to the {ggplot2} code you made for plotting the histogram which I think improved the plot that you were trying to make so you can further your plotting skills using the packaqge.
• When writing you code, always check if the code is running locally. If a line is erroring and you are not able to correct it, hash it out or delete it before going to another line to try something different. This is always good practice because when you run your code again, that line will error and can throw you off. Always remember that the code must work for it to stay on the script. If it doesn't work, correct it or hash it out or remove it.
• Creating objects is your friend. For you to be able to use the output of your code, you need to assign it to an object so you can use it later on. You sort of did this at the start but when you got to the part where you found the most appropriate approach to answer your questions, you defaulted again to not creating objects. You can see the difference of doing so in the updated code I made for your section in the RScript and in the R Markdown.
• It seems that you guys have forgotten how to write R Markdown. If you find the time, please review this topic as it is, in my opinion, the most accessible coding approach for beginners like you. To see a whole worked example of how to write an R Markdown report, see this discussion on GitHub - https://github.com/orgs/OxfordIHTM/discussions/7
• In your outputs, I think you forgot the additional grouping that was important/necessary when analysing the data you got. Remember that the data was for a baseline round and endline round (two time points of data). In your results, you just reported on the whole of the data without sub-grouping also by survey round. For our purpose of this exercise that is OK as we are learning to code but from an analysis perspective, your results are technically invalid as combining results for both time points doesn't really make sense.

Well done again! Congratulations with your participation in this hackathon. Look forward to the course dinner in April as we will be handing out awards and tokens for all of you guys.