BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
seamoh
Obsidian | Level 7

Hello all of my friends in the SAS community, I hope you are very well. I am doing NHANES tutorial Skip pattern, the link is below

https://www.cdc.gov/nchs/tutorials/nhanes/Preparing/CleanRecode/Task2.htm

this is about Skip Pattern, I read the tutorial but I don't get the Skip pattern concept. why BPQ020 BPQ030 shows a skip pattern, and why we recode the data?

Thank you very much

deeply appreciated.

I am looking forward to your answer

 

1 ACCEPTED SOLUTION

Accepted Solutions
ballardw
Super User

Skip patterns in surveys relate to whether a value should be there or not.

One simple example If a person is female you do not ask them male specific health questions, or you only ask questions about certain topics if the response to question is a specified answer, such as person that says yes to a question like "have you ever been diagnosed with diabetes" gets additional diabetes management related questions but someone that answers No would not have answers to those questions.

 

So to make the time for the interviewer and respondent quicker the questions are not asked at all, i.e "skipped". Which results typically in missing data based on the order and requirements in the survey.

I am not reading the entire website but I am going to guess that the (box 2) after the response categories is a note about which group of questions get asked next when that response is chosen.

Your specific question: BPQ020 is "Have you ever been diagnosed with high bloodpressure"(to paraphrase). If you answer this "NO" why would need to know BPQ030 "were you told 2 or more times"?

 

 

The "recoding" depends on what you intend to do with variables. You might want to insert a value for the missing variables that indicate "intentionally skipped" or remove the "don't know" and "refused" categories so you can more easily get the percentage of Yes/No responses among the respondents that answered one of those two categories.

Other recode approaches might reduce the number of categories because you had many choices, such as a 1 to 100 scale but the individual responses didn't provide much information but grouping the responses to "low" , "middle" and "high" with some rule might more easily show something of interest.

 

Another reason for recoding is providing and answer choice of "other" with a follow up to collect a description of what the respondent means by "other". Some times you can code that "other" back to a response category. One of the surveys I worked with asked question about what race the respondent identified as with response choices like "white", "black", "american indian", "asian" plus "other" and were asked what the meant by other. I have coded literally dozens of responses that were "white" and "black" entered in the "other" response.

 

View solution in original post

2 REPLIES 2
ballardw
Super User

Skip patterns in surveys relate to whether a value should be there or not.

One simple example If a person is female you do not ask them male specific health questions, or you only ask questions about certain topics if the response to question is a specified answer, such as person that says yes to a question like "have you ever been diagnosed with diabetes" gets additional diabetes management related questions but someone that answers No would not have answers to those questions.

 

So to make the time for the interviewer and respondent quicker the questions are not asked at all, i.e "skipped". Which results typically in missing data based on the order and requirements in the survey.

I am not reading the entire website but I am going to guess that the (box 2) after the response categories is a note about which group of questions get asked next when that response is chosen.

Your specific question: BPQ020 is "Have you ever been diagnosed with high bloodpressure"(to paraphrase). If you answer this "NO" why would need to know BPQ030 "were you told 2 or more times"?

 

 

The "recoding" depends on what you intend to do with variables. You might want to insert a value for the missing variables that indicate "intentionally skipped" or remove the "don't know" and "refused" categories so you can more easily get the percentage of Yes/No responses among the respondents that answered one of those two categories.

Other recode approaches might reduce the number of categories because you had many choices, such as a 1 to 100 scale but the individual responses didn't provide much information but grouping the responses to "low" , "middle" and "high" with some rule might more easily show something of interest.

 

Another reason for recoding is providing and answer choice of "other" with a follow up to collect a description of what the respondent means by "other". Some times you can code that "other" back to a response category. One of the surveys I worked with asked question about what race the respondent identified as with response choices like "white", "black", "american indian", "asian" plus "other" and were asked what the meant by other. I have coded literally dozens of responses that were "white" and "black" entered in the "other" response.

 

Reeza
Super User
Did you read through the NHANES survey? You should do that. Remember to not separate the coding aspects from the business problem you're trying to solve.

SAS Innovate 2025: Save the Date

 SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!

Save the date!

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

SAS Training: Just a Click Away

 Ready to level-up your skills? Choose your own adventure.

Browse our catalog!

Discussion stats
  • 2 replies
  • 1475 views
  • 5 likes
  • 3 in conversation