Mod Moved Comments To Chat
4 added 133 characters in body
source | link

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.


Based on a comment I made to the excellent answer of Amber B., an alternative to a bulleted list or an open-ended field and that is amenable to computer analysis is to create a binary tree of yes-no questions. Root is something like, for example: "Do you now identify as male?". Each answer, combined with all previous answers, is used to select the next question, say, for "no": "Have you ever self-identified as male?". It is complex as you suggest and implies a complete analysis that is tailored to your need and also sensitive to the understandings of others. Being complex, you need to justify your need for the information as you say (sociological, medical, vs curious).

The next answer, however, must be based on all previous answers, not just the recent one. It may be that "currently" identifying as male may not be enough, and that lots of combinations are possible. But this, at least gives you a way to analyze the results.

And I'll also note, based on my other comments there that: there are at least three independent factors in sex/gender. Your mind, your "plumbing", and your hormonal system. (I'm not a medical geek, but think that chromosomes have an influence on both plumbing and hormonal structure. All have an influence but none is absolutely determinative. In the past people discounted your mind - what you thought - and imposed a classification on you. Now, with modern medicine, surgery, and pharmacology, a person can bring the three to a self-desired balance. We now recognizes that people have choices as well as biology where this was denied in the past, and, by some, even now.

Finally (perhaps), none of those three factors is binary, not even chromosomes. All three have a range of values that combine into "who we are".

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.


Based on a comment I made to the excellent answer of Amber B., an alternative to a bulleted list or an open-ended field and that is amenable to computer analysis is to create a binary tree of yes-no questions. Root is something like, for example: "Do you now identify as male?". Each answer, combined with all previous answers, is used to select the next question, say, for "no": "Have you ever self-identified as male?". It is complex as you suggest and implies a complete analysis that is tailored to your need and also sensitive to the understandings of others. Being complex, you need to justify your need for the information as you say (sociological, medical, vs curious).

The next answer, however, must be based on all previous answers, not just the recent one. It may be that "currently" identifying as male may not be enough, and that lots of combinations are possible. But this, at least gives you a way to analyze the results.

And I'll also note, based on my other comments there that: there are at least three independent factors in sex/gender. Your mind, your "plumbing", and your hormonal system. All have an influence but none is absolutely determinative. In the past people discounted your mind - what you thought - and imposed a classification on you. Now, with modern medicine, surgery, and pharmacology, a person can bring the three to a self-desired balance. We now recognizes that people have choices as well as biology where this was denied in the past, and, by some, even now.

Finally (perhaps), none of those three factors is binary. All three have a range of values that combine into "who we are".

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.


Based on a comment I made to the excellent answer of Amber B., an alternative to a bulleted list or an open-ended field and that is amenable to computer analysis is to create a binary tree of yes-no questions. Root is something like, for example: "Do you now identify as male?". Each answer, combined with all previous answers, is used to select the next question, say, for "no": "Have you ever self-identified as male?". It is complex as you suggest and implies a complete analysis that is tailored to your need and also sensitive to the understandings of others. Being complex, you need to justify your need for the information as you say (sociological, medical, vs curious).

The next answer, however, must be based on all previous answers, not just the recent one. It may be that "currently" identifying as male may not be enough, and that lots of combinations are possible. But this, at least gives you a way to analyze the results.

And I'll also note, based on my other comments there that: there are at least three independent factors in sex/gender. Your mind, your "plumbing", and your hormonal system. (I'm not a medical geek, but think that chromosomes have an influence on both plumbing and hormonal structure. All have an influence but none is absolutely determinative. In the past people discounted your mind - what you thought - and imposed a classification on you. Now, with modern medicine, surgery, and pharmacology, a person can bring the three to a self-desired balance. We now recognizes that people have choices as well as biology where this was denied in the past, and, by some, even now.

Finally (perhaps), none of those three factors is binary, not even chromosomes. All three have a range of values that combine into "who we are".

3 incorporated comments
source | link

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.


Based on a comment I made to the excellent answer of Amber B., an alternative to a bulleted list or an open-ended field and that is amenable to computer analysis is to create a binary tree of yes-no questions. Root is something like, for example: "Do you now identify as male?". Each answer, combined with all previous answers, is used to select the next question, say, for "no": "Have you ever self-identified as male?". It is complex as you suggest and implies a complete analysis that is tailored to your need and also sensitive to the understandings of others. Being complex, you need to justify your need for the information as you say (sociological, medical, vs curious).

The next answer, however, must be based on all previous answers, not just the recent one. It may be that "currently" identifying as male may not be enough, and that lots of combinations are possible. But this, at least gives you a way to analyze the results.

And I'll also note, based on my other comments there that: there are at least three independent factors in sex/gender. Your mind, your "plumbing", and your hormonal system. All have an influence but none is absolutely determinative. In the past people discounted your mind - what you thought - and imposed a classification on you. Now, with modern medicine, surgery, and pharmacology, a person can bring the three to a self-desired balance. We now recognizes that people have choices as well as biology where this was denied in the past, and, by some, even now.

Finally (perhaps), none of those three factors is binary. All three have a range of values that combine into "who we are".

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.


Based on a comment I made to the excellent answer of Amber B., an alternative to a bulleted list or an open-ended field and that is amenable to computer analysis is to create a binary tree of yes-no questions. Root is something like, for example: "Do you now identify as male?". Each answer, combined with all previous answers, is used to select the next question, say, for "no": "Have you ever self-identified as male?". It is complex as you suggest and implies a complete analysis that is tailored to your need and also sensitive to the understandings of others. Being complex, you need to justify your need for the information as you say (sociological, medical, vs curious).

The next answer, however, must be based on all previous answers, not just the recent one. It may be that "currently" identifying as male may not be enough, and that lots of combinations are possible. But this, at least gives you a way to analyze the results.

And I'll also note, based on my other comments there that: there are at least three independent factors in sex/gender. Your mind, your "plumbing", and your hormonal system. All have an influence but none is absolutely determinative. In the past people discounted your mind - what you thought - and imposed a classification on you. Now, with modern medicine, surgery, and pharmacology, a person can bring the three to a self-desired balance. We now recognizes that people have choices as well as biology where this was denied in the past, and, by some, even now.

Finally (perhaps), none of those three factors is binary. All three have a range of values that combine into "who we are".

2 added 413 characters in body
source | link

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

If you are in the US you should (must) ask this question of your IRB. The answer may depend on the overall nature of your study, not just this one question. They will probably want to know how you intend to use this information (and everything else).

Rather than Male, Female, blank. It might be Male, Female, Decline to Answer. Even that will, however, offend some people who are, for example, non-binary. But if you think you need to list all possibilities you have quite a number of self identified genders. And you would have to justify the question. Some will consider the question intrusive for some purposes.

If you are not in the US, then local laws may also apply to any study with human subjects and you need to know this. I don't know the law in Germany.

One option that may be acceptable is to not use a bulleted list, but a field in which people can characterize themselves. This makes for a more difficult research design, of course, since you don't know in advance what the responses will be. Moreover, depending on who fills out the questionnaire, the results might be more creative than you like, as mentioned in the comments here.

Of course you don't know that people will answer honestly in any case. This could occur if you are asking for sensitive information otherwise.

Yet another complication for such research, that needs to be considered in the design. If you get, say 90% "male" responses, how to you make a determination about whether the results differ by gender? It's not impossible to do, of course, but you have to have a plan.

1
source | link