Hello, Selen Ozbek

1- For the first question, it depends on your data, you can compare the frequence(numbers) of Women in power ( I dont' knwo rraly the indicator you used for such phenomenon)

2- I'm not sure, but for this question, a SEM analysis will be more appropriate( because mindfulness is a proxy variable; precisily how we can quantify (or measur this variable??)

3- Yes for this question, you can use Multiple regression ; but the fact you have all your (IDv are categoricals) I suggest A tow-way ANOVA analysis.

4- Honestly, I dont' underttood well the last question , especialy the sentence " influence expectations of a leader in times of success versus crisis to contribute"

Best Wishes

Hi Selen Ozbek !

First of all, I think you're doing a very interesting research. Congrats! Regarding your doubts, I present the following suggestions:

1. How did you measure the glass cliff phenomenon? If you measured using Likert scale, then you should run a T Test (when the DV is continuous and the IV is discrete). If you measured it as a discrete variable, then you should run a chi-square test (and both the IV and DV are discrete variables).

2. Also depends on how you measured unconscious bias. Is unconscious bias discrete or continuous? If it is a continuous variable (and if you want to run a simple correlation test), I think a Pearson correlation test is enough.

3. If you manipulated both company performance and choice of candidate variables - and if you created four scenarios = 2 (crisis vs successful) x 2 (male vs female) -, then you should go for a two-way ANOVA analysis. But this is not clear in your question.

4. It depends on how you measured the expectation of a leader to contribute to facilitate glass cliff. If you measured this expectation with a Likert scale, and if the choice of candidate was manipulated, then you should also go for a two-way ANOVA, as you have two independent variables (choice of candidate and company performance) and one dependent variable (expectation of a leader to...). You'd do the same thing indicated in the point 3, but you'd change the dependent variable.

If you describe how you measured each of these variables, maybe I can help you more precisely.

Hi Reneta, many thanks for your reply.

Here is how all things measured:

variable1: residence, 2 levels (UK, Canada)

variable2: gender, 2 levels (male, female)

--randomization happened here--- half of the participants read a company history going successful, the other half read a company history going bad.

variable3: candidate choice (female candidate, male candidate) - both groups saw the same candidates.

variable4: question asked "the company will be able to solve their problems in the future" - measured with Likert scale with 7 options

variable5: question asked "the company will be successful in 10 years" measured with Likert scale 1 being bankrupted, 10 being one of the most successful companies.

variable6: mindfulness, measured with Likert scale and scored as means of the scale (1-7).

Thank you in advance!

Thank you, Selen! It is clearer now, but regarding question 1 ( Does glass cliff phenomenon differentiate between Canada and UK?) and question 2 (Is mindfulness correlated with unconscious bias to women in the recruitment process?), how you measured glass cliff phenomenon and unconscious bias still unanswered. Regarding question 1, if you measured glass cliff phenomenon as a discrete (or categorical) variable (e.g., yes/no levels), then I think you should run a chi-square test. If don't, if you measured it as Likert scale, then you should run a simple T Test for independent samples. Regarding question 2, if unconscious bias was also measured with a Likert scale (as mindfulness), then you should run a Pearson correlation test (if you objective is to analyze correlation and not

regression).

About question 3, I don't know if I can help you. I'm trying to solve this puzzle, because I understand you have a X - M - Y model, wherein X is the independent variable (crisis x successful scenario), Y is the dependent variable (future success perception) and M is the mediator (candidate choice - at list I see it as a mediator). I'm not sure if that's right. I've never run a model where the mediator is a categorical variable. I know it is possible, but I don't know how to do it. Maybe you should consult statistics specialists and see if they can help you.

About question 4, It depends on how you measured the expectation of a leader to contribute to facilitate glass cliff. If it was measured as Likert scale, then you should run a T Test.

glass cliff and unconscious bias are the same thing - I apologize about the wording. Glass cliff is the difference in candidate choice (people choose female if the company is going down and male if the company is successful) and my hypothesis was this is gonna occur with this population too and it will be correlated with mindfulness scores (i.e. those who chose male for successful company will be less mindful).

I did a google search and it shows multinomial logistic regression as I have 3 IVs and categorical DV. But it excludes mindfulness scale - would that be a bad idea to run a regression and seperate correlation to see the relationship with mindfulness?