I'm using sslope in stata to test the linear relationship between x and y for different values of my moderator. The output of sslope provides me with a significance level and a coefficient. As far as I know, a simple slope coefficient represents the slope of a regression line for a specific value of my moderator. I report standardized coefficients (betas, "b") in my regular regression table. However, I'm not sure how to report the coefficient provided by sslope. Should i denote this coefficient also "b"?
$\begingroup$
$\endgroup$
5
-
1$\begingroup$ What do you mean by "standardized coefficient"? $\endgroup$– DavidCommented Jul 8, 2019 at 10:01
-
$\begingroup$ betas as provided by the "beta" option in regression in stata. $\endgroup$– user18075Commented Jul 8, 2019 at 10:49
-
1$\begingroup$ You are mixing a lot of notation: simple slope coefficient / interaction / standardized coefficient. Which is it? $\endgroup$– user2974951Commented Jul 8, 2019 at 11:05
-
$\begingroup$ Indeed, I am confused. As far as I know, a simple slope is the slope of a regression line for a specific value of the moderator. The module "sslope" in stata provided me with this slope. My question is, how do I denote this slope/coefficient? $\endgroup$– user18075Commented Jul 8, 2019 at 11:34
-
$\begingroup$ Can you post your output? Are you just interested in how to write this up in a scientific paper? $\endgroup$– gung - Reinstate MonicaCommented Jul 8, 2019 at 20:00
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
I don't know what sslope is but it reminds me of linear regression, where we build a model $Y=\beta_0 + \beta_1 x + \epsilon$ where:
- $\beta_0$ is a constant.
- $\beta_1$ represents the slope of the line that best fits the data.
- $\epsilon$ is random noise
If this is the case, I would go for $\beta_1$ as the symbol for that coefficient. It there is no intercept (at $X=0$ we have $Y=0$ too) I would go for just $\beta$
