编程作业代写 Learning Objective Introduction to analysis

Learning Objective

编程作业代写 The grading for programming assignments will give you credit for doing the work, with some variation based on rubrics used appliedr

This assignment gives you practice integrating self-reported and measured information to classify disease status, in this case diabetes and applying population weights with adjustments for a complex sample design.

Assignment Description  编程作业代写

You will create, run, and test a SAS program file that describes diabetes status from NHANES 2017-2018. In your program you will compare self-reported diabetes (diagnosed) to Hemoglobin A1c (HbA1c, glycohemoglobin) measures to estimate the prevalence of undiagnosed diabetes and diagnosed diabetes under control or not. An individual with HbA1c that is >= 6.5 is considered to have diabetes. Those diagnosed with diabetes are considered to have it under control if HbA1c is less than 7.0. Assign diabetes status to each adult:

Run weighted tables, with standard errors adjusted for the sample design, of diabetes status overall and by sex, race-ethnicity, and education. Write a brief paragraph describing which groups appear to have higher rates of undiagnosed diabetes and uncontrolled diagnosed diabetes.

Variables you will need:  编程作业代写

1. Unique identifier (SEQN): needed for merging


2. Age: report estimates on adults (18 or older) only


3. Sex: male / female


4. Race-ethnicity: recoded to Hispanic, non-Hispanic white, non-hispanic black, non-hispanic other race


5. Education: recoded to less than high school, high school, some college or more


6. Whether the respondent says a doctor has told them they have diabetes.


7. HbA1c: lab value for HbA1c


8. Weight variables: population weight, PSU, and strata variables. Because labs are being used, the sample is restricted to those who have medical examination data. The appropriate weight variable is WTMEC2Y. Note that WTINT2YR is the weight for the survey portion which would be appropriate when using survey data only. You will also need SDMVPSU and SDMVSTRA.

Your results should show diabetes status overall, by sex, by race-ethnicity, and by education. You should use the weights with proc surveymeans. Note that you can use a CLASS statement with your diabetes status variable, also specified on the VAR statement, to produce percent in each category, e.g.:

CLASS diabetes_stat;
VAR diabetes_stat;

You should use the DOMAIN statement to run tables by other variables, e.g.: DOMAIN sex raceh educ;

You will also need CLUSTER, STRATA, and WEIGHT statements.

To get results into a SAS data set, you can use ODS, e.g.,: ODS OUTPUT DOMAIN=[file for results];

You can then export or use ODS HTML and PRINT to put the results into Excel.

The assignment is due 4/24/2021 at the end of the day.

Additional Notes

In this assignment you will use NHANES for 2017-2018. NHANES is a study conducted every two years, which collects survey information as well as results from a physical examination and blood tests.

Steps  编程作业代写

1. Identify the variables and files you’ll need:
   o Find the NHANES 2016-2017 documentation online
   o Locate the “Has a doctor ever told you that you have diabetes” question and the file it is in
   o Locate the file containing the HbA1c lab measure and the variable name
   o Find the race variable.
   o Find population weight, PSU, and strata variables.




2. Extract the relevant variables and merge the parts together. Note that the individual ID is SEQN.



3. Flag each individual as diagnosed diabetes under control or not, undiagnosed diabetes, and no diabetes


4. Run tables on the diabetes status variable for the full population and by sex, race, and education three ways with weights using adjustments to the standard error (PROC SURVEYMEANS in SAS). See notes above about specific statements to include (CLUSTER, STRATA, WEIGHT, DOMAIN, CLASS, and VAR). See also notes about getting results into a file using ODS.

Grading

The grading for programming assignments will give you credit for doing the work, with some variation based on rubrics used applied to your programs and their results.

There are 10 potential points.
Credit for doing the work
+3 for being complete and on time,
+4 for reviewing with the group

0 to 3 depending on rubrics for evaluating the programs and results of final team program:

1. Accuracy
   a. Do the variables reasonably reflect the original data?
   b. Have missing values been accounted for?
   c. How does the program check derived variables?
   d. Have the proper observations been included or excluded?
   e. Do the Ns (number of observations) make sense?
   f. Have the right results been shown?

2. Program organization:

a. Is the program easy to read and understand? Could another person discern the
steps it’s taking, i.e., figure out what it’s doing?
b. Are variable names meaningful or are they labeled? Are the meanings of
categorical variables clear?

1. Efficiency:

a. Are files saved of reasonable size?
b. Are there unnecessary reads and writes of files?
c. Are there unnecessary processing steps?

更多代写： HomeWork cs作业     金融代考    postgreSQL代写         IT assignment代写     统计代写  数学Assignment代写