delivered the opinion of the court:
Plaintiff, United Air Lines, Inc., filed a complaint in the circuit court of Cook County for review of a final order entered by the Illinois Fair Employment Practices Commission finding that United had committed unfair and discriminatory employment practices. The trial court upheld the Commission’s action and United appeals.
On December 18, 1973, Paulette Julian filed a charge of unfair employment practice with the Commission alleging that United discharged her as a probationary flight cabin attendant as part of a concerted effort by United to prevent her and other black flight *520attendants from attaining union status. The pertinent evidence relating to Julian’s discharge was adduced at a hearing before a Commission hearing examiner.
On October 14, 1973, the senior flight attendant on a United flight reported to the stewardess supervisor at O’Hare Airport that Julian had not occupied her assigned seat during landings and takeoffs and had disregarded her instructions to occupy the seat on this flight and on other occasions. The attendant also informed the supervisor that Julian had placed United property in her suitcase and had taken it from the aircraft. Julian admitted to the supervisor that she had not occupied the seat but denied taking United property. With Julian’s permission, the supervisor searched her suitcase and purse and found the following United property: cartons of orange juice, a package of napkins, packets of sugar and instant coffee, and a writing portfolio. Julian then admitted that she placed the items in her luggage and that she knew removal of United property was grounds for immediate dismissal. The following day the supervisor prepared a report for the chief supervisor recommending Julian’s discharge. Before doing so, she reviewed Julian’s record and noted three letters of praise from passengers. The supervisor had not discharged any employee previously for taking United property. She had discharged white attendants in the past, but previously had not disciplined a black employee. The supervisor testified that the decision to discharge Julian was not racially motivated.
Julian testified that she was a probationary employee at the time of her dismissal. There was no animosity between her and the senior flight attendant. She had various items left over from the flight and had intended to use them or return them to the aircraft.
Marsha Hunter testified for Julian that she had observed a white flight attendant, Bonnie Fisher, carrying United liquor miniatures in her purse and had reported it to her supervisor two weeks later. No action was taken.
Pat Christian, Bonnie Fisher’s supervisor, testified that she talked to Fisher about the liquor after receiving the complaint. Fisher denied the accusation and, since she received the report two weeks later, nothing could be done.
Herbert Jackson testified that he promulgated the appropriate sanctions in behalf of United for its employees’ failure to comply with company rules. The company rule for theft was immediate discharge. That sanction had been applied to both black and white employees and he knew of no instance where it had been applied in a discriminatory manner. Jackson also knew of no United policy to eliminate black flight attendants.
After hearing the evidence, the examiner issued a recommended *521order dismissing Julian’s complaint with prejudice. He found that race was not a reason for Julian’s dismissal.
Thereafter, Julian filed a petition for a hearing de novo with the Commission. The Commission subsequently remanded the cause to an administrative law judge to take additional evidence. In its remandment order the Commission found that the evidence established a prima facie case of racial discrimination against United. It also found that the evidence established a legitimate basis for Julian’s dismissal and that Julian had not demonstrated that white attendants similarly situated were treated differently by United. The matter was remanded solely for the taking of statistical evidence to ascertain whether the basis of Julian’s dismissal was pretextual.
On March 17, 1976, statistical data and analysis were presented to a new administrative judge, the original hearing examiner having died. On September 27, 1976, the judge issued a recommended finding that the basis for the dismissal, according to statistics, was pretextual. The statistical data as reflected in the recommended finding covering the years 1968 through 1973, was as follows:
“1. Of the 8,268 white and black flight attendants entering employment with United the racial composition was as follows:
Number Percent
White 7817 94.5
Black 451 5.5
Other Minorities Unknown
2. Systemwide, 180 white flight attendants, 2.3% of their number entering, were discharged by respondent. By comparison, 40 black attendants, 8.8% of their number entering, were discharged by respondent. Thus, it was nearly 4 times more likely that a black flight attendant entering employment would be discharged than that an entering white flight attendant would be discharged. The statistical probability of this occurring by chance is infinitestimal, or .000. (amended, Sept. 28, 1976)
3. Systemwide, 133 white flight attendants, or 1.7% of their number entering, were discharged before the end of the six-month probationary period. By comparison, 35 black flight attendants, or 7.8% of their number entering, were discharged before the end of the probationary period. Thus it was approximately 4% times more likely that a black flight attendant entering employment would be discharged prior to the end of her probationary period than that an entering white flight attendant would be so discharged.
4. Among both white and black flight attendants a discharge was *522likely to take place before the six-month probationary period ended rather than after:
Black Systemwide O’Hare Base
Number discharged before end of probationary period 35 20
Number discharged 40 21
Percent 88% 95%
White
Number discharged before end of probationary period 133 75
Number discharged 180 105
Percent 74% 71%
The statistical probability of the greater percentage of black flight attendants being discharged before the end of probation occurring by chance is .067, which means, according to the expert testimony, it could have occurred by chance.
5. Respondent’s data show that 5 of 41 (12.2%) black flight attendants were discharged for violation of company rules while 50 of 241 white flight attendants (or 20.7%) were so discharged. The statisticians for both parties agree, however, that this configuration could have occurred by chance; that is, it is not a reliable statistical fact.
6. The parties were unable to determine facts comparable to those in Finding Nos. 2 and 3 for O’Hare base because respondent’s records did not reflect the base at which a flight attendant entered employment. Complainant’s expert testified without contradiction, however, and I find, according to statistical probabilities based on other calculations reflecting a similarity between O’Hare base and systemwide, the rate of discharge from O’Hare base was approximately the same as that rate systemwide.”
On March 14,1977, in a 3 to 2 split decision, the Commission adopted the findings of the administrative law judge that the discharge of Julian was pretextual. Additionally, the Commission supported its finding of pretext from a consideration of the trivial nature of the offense, the timing of her discharge, and her previous good work record.
Under the Administrative Review Act (Ill. Rev. Stat. 1975, ch. 110, par. 264 et seq.), the findings and conclusions of the administrative agency *523on questions of fact are to be held prima facie true and correct. (Northern Illinois University v. Fair Employment Practices Com. (1978), 58 Ill. App. 3d 992, 374 N.E.2d 748.) It is not the function of the reviewing court to reweigh the evidence or to make independent determinations of fact. (General Electric Co. v. Fair Employment Practices Com. (1976), 38 Ill. App. 3d 967, 349 N.E.2d 553.) We must decide only whether the agency’s determination is against the manifest weight of the evidence. (Unger v. Sirena Division of Consolidated Foods Corp. (1978), 60 Ill. App. 3d 840, 377 N.E.2d 266.) For a judgment to be against the manifest weight of the evidence, it must appear that conclusions opposite to those reached by the trier of fact are clearly evident. King v. City of Chicago (1978), 60 Ill. App. 3d 504, 377 N.E.2d 102; Mobley v. Conlisk (1978), 59 Ill. App. 3d 1031, 376 N.E.2d 247.
Statistical analysis serves an important role in cases of employment discrimination, and statistics are competent proof of such, but, like any other kind of evidence, their usefulness depends upon all the surrounding facts and circumstances. Hazelwood School District v. United States (1977), 433 U.S. 299, 53 L. Ed. 2d 768, 97 S. Ct. 2736; International Brotherhood of Teamsters v. United States (1977), 431 U.S. 324, 52 L. Ed. 2d 396, 97 S. Ct. 1843.
In our view, the statistical data and analyses offered in the present case do not support a finding that Julian’s dismissal by United was a pretext for a racially discriminatory policy. All that was shown was a statistical disparity between the percentage of black and white flight attendants discharged. No statistically reliable pattern of discharging black flight attendants by United was evident. The statistics did show that black flight attendants were dismissed 4% times more often than white attendants during the six month probationary period. No evidence was introduced that white flight attendants accused of theft received less severe sanctions than those imposed on black attendants. The data did indicate that more white attendants were dismissed for rules violations than were black attendants. There was no showing as to why either the black or white attendants had been discharged. We believe that, under the facts and circumstances of this case, the statistics bore no relationship to Julian’s dismissal. Even the Commission recognized that the statistical data was insufficient to support its findings and sought to bolster it by considering Julian’s prior good work record and the triviality of her offenses. The evidence was clear and overwhelming that Julian was discharged for theft, and that she was aware of the penalty for that offense. The evidence also was overwhelming that her race had nothing to do with her discharge. We conclude, therefore, that the decision of the Commission was contrary to the manifest weight of the evidence.
*524For the reasons stated, the judgment of the circuit court of Cook County is reversed.
Judgment reversed.
McGILLICUDDY and RIZZI, JJ., concur.