Hello all,   I have a dataset with 1000 observations and 6 variables (below is an example). The values of the variables bev_1-bev_5 are open ended text and contain different spelling variations. I'm creating additional variables which classify the responses in each bev_1-bev_5 column as 1=drink or 2=food and the type of drink. My goal is to set up a lookup table or essentially another dataset which includes all of the values of bev_1-bev_5 and the corresponding codes for my new variables. This lookup table will also have to include the spelling variations (e.g., Coke, Coca-cola, CocaCola, coke, etc. should all be coded as 1 for my product_Type variable, and should all be coded the same for any other variables I want to add). Example dataset: ID bev_1 bev_2 bev_3 bev_4 bev_5 1 Coke Starbucks Pepsi CocaCola fries 2 Coca-Cola burger pepsi coke redbull   Question: I'm wondering what the best way would be to structure the lookup table so that for example, anytime SAS sees the word "Coke" or any of its variations (which I would include in my lookup table) in bev_1-bev_5, it will code it appropriately based on the new variables I'd like to add.   Lookup table that I want to link, essentially as a "dictionary" to my original dataset. Columns are the new variables I'll be adding: Response and variations product_type type of drink Coke, Coca-cola, Cocacola 1 1 burger 2 0 redbull, Red Bull, RedBull 1 3 ... View more

Hi all,   I have 5 variables (bev_1-bev_5) that are open text responses. I am trying to create a new variable for each bev_1-bev_5 which classifies the responses in each column as either 1 or 2. The first new variable that I want to create, called product_type  should have 2 levels: if "Coca-Cola" or "Pepsi" are found in bev_1-bev_5 then the the variable product should =1. If "Hamburger" is found in any observations of bev_1-bev_5 then product should =2.   An example of my dataset is below. Variables product_1-product_5 are the new variables I am trying to create. Product_1 classifies the responses under bev_1, product_2 classifies responses of bev_2, etc.   ID bev_1 bev_2 bev_3 bev_4 bev_5 product_1 product_2 product_3 product_4 product_5 1 Coke Coke Pepsi Hamburger Hamburger 1 1 1 2 2 2 Coke Hamburger Pepsi Hamburger Hamburger 1 2 1 2 2 3 Pepsi Pepsi Pepsi Hamburger Coke 1 1 1 2 1   Here is an example of the progress I've made so far with my code. This code runs but ouputs only one variable and identifys which column "Coke" or "Pepsi" is located.  I'm also not sure if an array is the most appropriate function to use in this case. Any help is appreciated! data want; set have; array x{*} bev_1-bev_5; product_type=whichc("Coke","Pepsi",of x{*}); run;   ... View more

Hello all,   I have a dataset with 5 variables and I am trying to sort my observations alphabetically across all 5 variables. Is this at all possible in SAS?       ... View more

Hi all,    I have a dataset (see example below) of the number of social media posts by platform type and account type. I've run a Fisher's exact test which has turned out to be significant and now I want to determine where the actual differences in proportion lie. For example, I'd like to determine if the proportion of posts by Account 1 on Platform1 are significantly different in comparison to all other account types. I've performed a post hoc test where I made a new dichotomous variable, 1=Account 1, 0=all other accounts and then run a Fisher's exact on this 3x2 table. EXAMPLE TABLE: Frequency of Posts   Account1 Account2 Account3 Account4 Account5 Platform1           Platform2           Platform3           QUESTIONS: 1) I'm wondering if based on my aims, it's correct to group the columns as I have for a post hoc test. Most examples of pairwise comparisons as post hoc tests that I've seen tend to group the rows rather than the columns.  EXAMPLE PAIRWISE COMPARISON   Account1 All other accounts Platform1     Platform2     Platform3     Bonferroni p<0.0001 2) From my output, I get 5 p values (1 per each account type). For example, if the p value of my first pairwise comparison (see example above) is significant, then does this mean that there is a significant difference in the proportion of posts for Account1 compared to all other accounts for each of the 3 platforms? Is there a way in SAS to determine which exact cells within one row are significantly different between account types? ... View more

Hi everyone,   I have a conducted a Fisher's exact test with Monte Carlo simulation which resulted in a significant p value. My row variable, which is the # of people, has 4 levels (0, 1-50, 51-100, 101-150) and my column variable has 5 levels (AccountA, AccountB, AccountC, AccountD, AccountE). To identify the significant pairs between groups, I've conducted multiple comparisons using a 2x5 tables (e.g., 0 vs 1-50, 0 vs 51-100 etc.) with Bonferroni adjustment, and the adjusted Bonferroni p values are below. I am unsure of how to interpret the Bonferroni p values...for example, how would the adjusted p value for Test 1 (0 vs 1-50 people) be interpreted?      ... View more