Creating a scale with dichotomous and continuous items

[PHD_Student]

Hi all,

I am creating an acculturation scale (I know, fraught territory) from three NHANES items -
Nativity (dichotomous)
length of time in US (continous)
language (I am treating as continuous)

I am not sure the best way to standardize or transform these indicators to create a scale and haven't been able to find anything in the literature that does something similar. I've thought about mean centering vs Z-score and admittedly am struggling to understand what is happening under the hood with SD that would make one method better than the other.

I have also considered LCA but am currently being advised against it by my advisor as this would cut my sample in a way that wouldn't make theoretical sense with my question.

Thank you,

 

[AngleWyrm]

Check out Principal Component Analysis

 

[Karabiner]

Check out Principal Component Analysis

He will then learn in a second that this requires similar scale levels for all items, and cannot
be done with binary variable, no?

 

[Karabiner]

Hi all,

I am creating an acculturation scale (I know, fraught territory) from three NHANES items -
Nativity (dichotomous)
length of time in US (continous)
language (I am treating as continuous)

Could you tell us something about the topic and the research questions of your study,
and about the sample and the sample size, and what you will further use that scale/score
for?

With kind regards

Karabiner