Zsolt Sándor and Philip Hans Franses, "Consumer Price Evaluations through Choice Experiments", Journal of Applied Econometrics, Vol. 24, No. 3, 2009, pp. 517-535. The data set consists of choice data on hypothetical mobile phones collected from students of the Erasmus University Rotterdam in November-December 2003. The data set consists of four files: des.asc, freq.asc, nobs.asc, and qrdraws.asc. All four files are zipped in the file data-sf.zip. Since they are ASCII files in DOS format, Unix/Linux users should use "unzip -a". The file DES.ASC has 480 rows and 7 columns, where each row contains the characteristics of one hypothetical mobile phone. Each group of 24 rows are one questionnaire design consisting of 12 choice sets of two choices of hypothetical mobile phones. A respondent was given one such questionnaire design to make 12 choices; there are 20 such different questionnaire designs altogether. This way, this data file consists of 240 choice sets altogether; the frequencies that the first alternative of a choice set is chosen are contained in the file FREQ.ASC. This way, this file has 240 rows and 1 column. The coding of the characteristics is explained in Table A1 of the paper. The first 6 columns are the 6 characteristics from Table A1 in the same order. For the 3rd characteristic, Extras, 1, 2, 3 from the data file correspond to (1 0), (0 1), (-1 -1), respectively, from the table. Column 7 contains the price complexity dummy variables: It is 0 if the price of the hypothetical mobile phone is specified directly and it is 1 if the price is specified with discount. The file NOBS.ASC contains the number of respondents for the different questionnaire designs. This way, it has 20 rows and 1 column. The file QRDRAWS.ASC contains a quasi-random draw of normal random vectors used to estimate the random coefficient logit probabilities. It has 343 rows and 7 columns, of which the first 6 columns (i.e., 6 dimensions) are used for Model (1) and the first 3 columns (i.e., 3 dimensions) are used for Model (2), (4) and (5).