I'm currently going nuts with my data at the moment and I'm in a real pickle! Basically I ran a reading task that asked people to read sentences a word at a time, then they were asked a comprehension question to see if they understood what they read. The reaction times all the words and questions were recorded and inputed into SPSS.
I entered the data in following format:
1. sentence number
2. item number (i.e. the word or the question)
3. item reaction time (again this is for either the word or question)
4. question (whether they answered the question correctly or not)
Now here's my problem. I want to ask SPSS to group the sentences and questions together but using the sentence number to group them together. Then I want SPSS to eliminate those groups in which the questions were answered incorrectly. The problem lies in the fact that the variable entered for 'question' would only be present next to the item in which the question is found/answered. Therefore the lines that represent the items in which words are shown, in the column for 'question' there wouldn't be a variable, I hope that's not too confusing. Suffice to say next to the item which is a word, there is a missing value in the question column.
What was suggested and did occur to me was grouping the data and recoding the missing variables so that the entire sentence would mirror the code in the question box as the answer in the question item itself. I just don't know how to do that because I seem to be hitting loads of dead ends.
I'd appreciate any help I can get.