Zarni Lynn Kyaw

Dear Tanyawat,

Your CRF is very detail and information rich and I like it but here are a few points you can use to improve upon it.

1) Eligible Criteria – I also was thinking about displaying full Eligible Criteria in my CRF but in order to reduce redundancy, some of the variable can be reduced. e.g. Aged between 18 and 60 years old on the day…[Read more]

Dear Preut,

Thanks for your insightful reply. Great minds do think alike, some of your comments are same as Arjan to improve my CRF.

1) I was trying to choose between EU formant and The Good Clinical and Data Management Practice (GCDMP) and settled on MM-DD-YYYY but you are absolutely right. I will use DD-MM-YYYY from now on.
2) Yes, I will try…[Read more]

Arjan do we need to put up our CRF here on our own? I was waiting to review my assign CRF but it is sill not available.

Data standardization enables the rapid design, build, analysis, and submission of clinical trials. This means new treatments, procedures and tools get to market more quickly, for less cost. Data standards also support semantic interoperability, which is the ability to fully understand the meaning of data across different systems and sources. This…[Read more]

I would like to showcase two projects that I have worked on in my experience.

1) The People’s Health in Southeastern Myanmar: Results from a Household Survey and the Way Forward where we collect data from 486 households
2) Ethnic Health Organizations in Myanmar Readiness to provide Essential Health Services and Plans of Action where we c…[Read more]

– Steps in the data management workflow:

A data management workflow for a research project consists of six steps: planning and design, collect, generate and store, clean, analyze and visualize, manage, store and preserve, share and publish, discover, reuse and cite. We have done some of these steps in our household survey, such as planning and…[Read more]

I would like to share about a bi-annual study we did a few years back called : The People’s Health in Southeastern Myanmar: Results from a Household Survey and the Way Forward

– Purpose of data collection: The purpose of data collection was to monitor and evaluate a strategic purchasing project that aimed to improve health service delivery and c…[Read more]

Thanks Arjan for pointing it out. Well noted and I will not over emphasize on one sample’s results.

By the way, you can use https://snipboard.io/ to share the infographic directly here. https://prnt.sc/ doesn’t work here because it doesn’t give a link ending with .jpg, but snipboard do and I use . 🙂

I also used Canva, this is a great infographic. Thanks for sharing it with us.

Your explanation is direct and easier to understand than what is written in the text book but let me put my thoughts here as well 🙂
We cannot say that an effect size outside the confidence interval is wrong or impossible, because the confidence interval depends on the assumptions we make and the level of certainty we choose. Even if the effect…[Read more]

Yes, it seem our cohort all seems to think this is a big revelation after reading the whole thing. We read so many papers that just accept P ≤ 0.05, and not share the exact P value.

I read this article An unhealthy obsession with p-values is ruining science in Vox and it was a well written piece.

Wow, you actually include multiple points and I liked that cheers.

Nicely written, I agree, the best way to compare groups or studies is to use a test statistic and a P value that directly measure the difference and the chance of seeing it by chance.

I agree with both you and Farkhan point, the only exception is when the P value is very small (e.g., under 0.001), because then it does not matter much and it might be hard to calculate accurately.

You point is salient and well written, may I chime in?
P value is the chance of seeing a result as extreme or more extreme than the one observed, assuming that the test hypothesis is true. A large P value does not mean that the test hypothesis is true or that there is no effect. It only means that the test hypothesis is compatible with the data,…[Read more]

Please let me chime in with my understanding of this point as well.
Two-sided P value considers both positive and negative differences. One-sided P value considers only one direction of the difference. The type of P value depends on the type of question we are asking. For example, if we want to know if a new drug is better than an old drug, we use…[Read more]