I am trying to decipher the validity and reliability of a pre existing survey? Here ia what the article from the researcher states:
Sample size and statistical methods
No formal sample size calculations were performed. Based on considerations of feasibility, we aimed to collect data from 400 to 600 participants. We used descriptive statistics to summarise sample characteristics and Pearson’s chi-squared test to measure the association among pre-selected categorical variables. For the latter analyses, we applied a Bonferroni-adjustment to account for multiple testing (alpha level at 0.005). Additionally, we used multivariable logistic regression to identify predictors of having a desire to use PrEP or a history of PrEP use. Odds ratios and their respective 95% confidence intervals were used to quantify the effects. To select variables for our multivariable model, we compiled the following initial working set of potential predictors in which we had a priori interest based on background knowledge: age, financial situation, education, family origins, sexual risk behavior, self-perceived risk, having peers living with HIV, and perceived barriers and risks of PrEP. For pragmatic reasons of reporting and traceability, we subsequently screened these using simple (i.e., univariable) logistic regression and included in the multivariable model those variables that were associated with the respective dependent variable at a p-value cut-off point of 0.075 following the approach described by Bursac et al. [39]. We later conducted a sensitivity analysis with all variables of a priori interest to ensure that important adjustment variables had not been overlooked. Missing cases were excluded in a listwise fashion.
To avoid collinearity of independent variables related to different measures of sexual risk behaviour in our logistic regression models, we created a new variable comprising four groups as shown in Table 1. In doing so, we aimed to approximate roughly the indications for PrEP use recommended by the CDC for MSM. We chose “two or more partners” rather than “one” as our cut-off point to account for the possibility that participants who reported condomless anal intercourse (CAI) with one partner might be describing CAI within a monogamous partnership. We did not distinguish between receptive or insertive CAI because the CDC indications for PrEP use for MSM do not do so either.
I am utterly confused. Any help would be appreciated.