Meta-analysis

Posted 2020-11-26 oareii

When I am going throught the first revision of my first meta-analysis about executive funtion in ASD group, I have to start the whole process over for answering the reviewers questions, and I am questioning myself:

 

1. When I want to check the definition of "fluency", it‘s really a nightmare, it is said that "...verbal fluency requires executive function such as "the ability to initiate and maintain effort and organize information for retrieval"......including selective and sustained verbal attention, vocabulary knowledge, storage and retrieval of long-term semantic and lexical knowledge, and aspects of executive funtion such as strategic search and switching...", then I looked through wiki, it is said that EF include basic cognitive processes such as attention control, cognitive inhibition....."

%%

I think I ruled out studies that only studied "attention" cause I thought it doesn‘t belong to EF, but for now, it‘s ok since I said it clear in my paper that I only examined five key components in EF other than all domain... so...

*next time, I should make myself prepared about what I am going to study, it‘s not  acceptable at all as a scientific research actually.

 

2. It‘s about the principle of inclusion and exclusion. It suddenly occured to me that I included some results of WAIS, but actually every study which focused on adult group with HFA may reported the subscore of WAIS...

%%

In a word, there never is a principle or standard for the study selection, it‘s may induce a lot of subjective bias.

*firstly, meta-analysis doing by two person may be more reasonable since bias can happen in every stage; secondly, analyses should be scrutinized, if it‘s not possible for having another senior doing this, it maybe helpful to leave it aside for a while before double-checking.

 

3. However I try I could not get the same result in meta-regression...I totally have no idea how I got these results......god damn it......

*Every step of analysis should be recorded!!! The same is true of writing.