The current theoretical analyses of the parity nonconserving (PNC) experiments in light nuclei fail to provide a consistent description of the experimental data for ' F, ' F, and 'Ne. We perform a "piecewise" analysis of the PNC matrix elements in the Op&/2-Od5/2-1sl/2 model space and show that by using a proper overall normalization of the isoscalar and isovector parts we can consistently describe all three PNC experimental data. New calculations are also presented for the s p sd p-f -mo-del space which support our analysis. PACS numbers: 24.80.Dc, 11. 30.Er, 21.60.Cs, 27.20.+n The investigations of low energy parity nonconservation (PNC) phenomena in light nuclei have as a goal to provide more reliable results for the hadron-meson weak coupling constants. These couplings are of importance for our understanding of the quark behavior inside the nucleons under the influence of the fundamental interactions. These investigations necessitate both very delicate experiments and very reliable nuclear structure calculations of the matrix elements for a correct extraction of the weak nucleon-meson coupling constants. Most of the results on the experimental and theoretical PNC studies in light nuclei have been presented in a review paper [1]. From the proposed cases during the last 25 years in this range of nuclei, four cases have been selected as reliable enough for experimental and theoretical analysis. They involve parity mixed doublets (PMD) [1] in ' N, 'sF, '9F, and 2'Ne. Two others cases involving PMD's in '60 [2] and 2oF [3] have been proposed recently. From the four mentioned cases, only the ' F has been measured with a result larger then the experimental error. All other cases have been measured with errors larger ('sF and 2'Ne) or near the result ('4N).However, the upper limits obtained for '~F and 'Ne are so small that they impose severe constraints on the different contributions to the PNC matrix elements. These results, compared to current theoretical calculations, have shown a discrepancy, which has not yet been solved (see also Fig. 11 from Ref. [4]). Namely, if one interprets the small limit of the (experimentally) extracted PNC matrix element ((0.029 eV) for 2'Ne as a destructive interference between the isoscalar and the isovector contribution [1], then it is difficult to understand why the isovector contribution in 's F is so small ((0.09 eV) and the constructive isoscalar + isovector contribution in ' F is relatively large (0.40~0.10). In the last few years, efforts have been made to improve the shell-model calculations with special emphasis on the description of the weak observables [5,6]. Recently, two new interactions have been developed by Warburton and Brown [7], which were designed to accurately describe the energies of states in nuclei with A = 10 -22. The interaction was designed for pure oscillator excitations [e.g. , OIico (Op)" configurations, 1Itco (Op)" '(Odls)' configurations, etc. ], but methods have been developed to use them when mixed n Fico excitations contribute [6,8]. Recently, w...