While the correlation between the CAG repeat length of the androgen receptor (AR) gene and idiopathic male infertility is still unclear, ethnic background of the population studied may play an important role in this association. The objective of this study was to determine whether changes in the CAG repeat length are associated with spermatogenic defects in Turkish infertile men. Reproductive hormone concentrations and the CAG repeat length in exon 1 of the AR gene in 47 idiopathic infertile men and 32 fertile controls were analyzed. The mean serum luteinising hormone (LH) and follicle stimulating hormone (FSH) levels were significantly higher in the infertile group than those of the control group (p < 0.0001 for both comparisons), whereas the mean serum testosterone level in the infertile group did not differ significantly from that in the control group (p = 0.16). The mean CAG repeat length of the AR gene in the infertile group did not differ significantly from that in the control group (22.28 ± 0.37 vs 22.41 ± 0.54, respectively; p = 0.84). In addition, the frequency of having a CAG repeat number ( 24) was also comparable between the infertile patients and fertile controls (31.9% vs 40.6%, respectively; p = 0.21). In conclusion, reproductive hormones with elevated LH and FSH, and normal or low testosterone levels were suggestive of partial impairment of testicular function. However, no statistically significant relationship between the length of the CAG repeat and idiopathic impaired sperm production was observed in the Turkish population studied. These results support the findings of previously published European studies, but are contrary to the findings from Caucasian and North American population studies. Thus, ethnicity and genetic backgrounds seem to be important in this association, and studies from a variety of different ethnic and genetic backgrounds using