To interpret the multiple regression results, you should first look at the R squared which represents the percentage that changes in the independent variables (coefficients) effect the dependent variable (null hypothesis). Next, you you check at what level of confidence the result is statistically significant at using the F statistic. These are the basic tests; however more include multi-colinearity and auto-correlation.
Multiple regression analysis assumes that data is numerical in nature (quantitative). If your input data is qualitative (such as Questions and Answers which I have interpreted 'QA output' as) then it is necessary to convert the data into numerical equivalents that can then be used as predictive components within the regression equation. For example, if the the answer can be dissatisfied, neutral or satisfied then this can be converted to a scale of 0-2 to represent the level of satisfaction.
