Missing data is an ever-growing issue in the modern-day world. Running critical analyses on incomplete datasets is a burden; if not all, fields must deal with it at some point. This paper explores several studies on Rhinoconjunctivitis Quality of Life (RCQOL) questionnaires to prove the effectiveness of tensor-based techniques, particularly four-dimensional and/or binary tensors, for handling missing data over more traditional ones used in mainstream research. Each missing value was predicted based on its correlations with known values by converting all decimal numbers to binary. This enables the program to predict missing values and calculate the probability that each possible value has to replace a missing one, which is particularly important when conducting cohort analysis. Furthermore, this paper reports on the setup and testing of seven different methods for the handling of missing data, with the addition of the extra binary dimension, on the basis of a dataset comprising the responses of 43 patients.