While the t-test works on continuous variables (e.g. body temperature, or age), Chi-squared (pronounced "kai") applies to categorical variables. There are two scenarios for using the Chi-squared test:
The first one is where you have one sample, and you know the proportions of the variable in the general population.
For example, you see 60 female patients and 40 male patients ("your sample"), but the overall female-male proportion in the catchment area is 55%:45%. You can then use the Chi-squared goodness of fit test (also known as "Chi-squared for given probabilities", or "One-sample Chi-squared") to tell you how likely it is that the difference you are seeing (you see a 5% higher proportion of female patients) is purely by chance.
The second option compares the observed counts from two or more groups directly, so not with a known distribution like the general population.
The underlying computation, as well as the interpretation of the two versions is similar, and in most cases is just referred to as the "Chi-squared test".
First question: Is your sample statistically significantly different from the general population?