Dear Stata-Community,
I'm currently in the process of analyzing data for an online survey my group designed and conducted. We want to investigate if extreme protest behavior reduces support for a social movement from the perspective of an outsider. Each participant was randomly assigned to one of four groups in which they read a news transcript about a protest action. Two of the four groups read a news transcript in which the protest was meant to be moderate (either about animal rights or airport extension); the other two groups had transcripts where the protest strategy was meant to be extreme (also either about animal rights or airport extension).
In the exported CSV-file, the missing fields (those to which the participant did not get randomly assigned to) have a value of -77 (see image 1). If I try to work with the observations for which there is a value of 1 to 5 (each item could be rated from 1 Not at all to 5 Extremely) measuring if they support the group they have read about, then Stata automatically excludes those values since the same column also has an observation of -77.
I would like to run some regressions but that is not working due to each column also includes the value -77 which I cannot get rid of without deleting all values within that same column. Note that one column is one variable..
I understand that my description is really poorly formulated as I lack precise wording and theoretical knowledge of inferential methods. I do, however, hope that there is someone who understands the issue and can point me in the right direction.
Many thanks in advance for taking your time.
Best wishes,
Nicholas
I'm currently in the process of analyzing data for an online survey my group designed and conducted. We want to investigate if extreme protest behavior reduces support for a social movement from the perspective of an outsider. Each participant was randomly assigned to one of four groups in which they read a news transcript about a protest action. Two of the four groups read a news transcript in which the protest was meant to be moderate (either about animal rights or airport extension); the other two groups had transcripts where the protest strategy was meant to be extreme (also either about animal rights or airport extension).
In the exported CSV-file, the missing fields (those to which the participant did not get randomly assigned to) have a value of -77 (see image 1). If I try to work with the observations for which there is a value of 1 to 5 (each item could be rated from 1 Not at all to 5 Extremely) measuring if they support the group they have read about, then Stata automatically excludes those values since the same column also has an observation of -77.
I would like to run some regressions but that is not working due to each column also includes the value -77 which I cannot get rid of without deleting all values within that same column. Note that one column is one variable..
I understand that my description is really poorly formulated as I lack precise wording and theoretical knowledge of inferential methods. I do, however, hope that there is someone who understands the issue and can point me in the right direction.
Many thanks in advance for taking your time.
Best wishes,
Nicholas
Comment