Ere found to be significant in the model (indicated by asterisks accompanying the Wald statistics). The decision was made to use a logistic regression and Odds Ratios (ORs), rather than prevalence rates or relative risk (RR), following suggestion [32, 33] that in non-controlled sampling cases where there is high outcome prevalence (above 10 ), ORs are a more appropriate summary measure with a more tenable assumption of homogeneity across all individuals in a sample population. Note therefore that the estimates should not be interpreted as approximated risk. The model (p < .001, 2 [24, N = 473] = 165.97, Nagelkerke R2 = 40 , correctly predicted cases = 74.2 ) indicated that only three factors--grade level, school, and whether or not one's sibling had a certificate--remained significant predictors. Odds ratios indicated that students in grade eight had significantly higher odds of possessing a birth certificate when compared to lower grade levels, while having a sibling with a certificate was associated with higher likelihood of the student him/herself possessing a certificate.Students' suggestions for improving certification of their own birthsStudents' own ideas for improving certification rates are reported in Table 4. Regarding how the government could improve registration, student answers--which were given in a free style with no prearranged answer choices--showed a rather even distribution between the themes of (1) direct action, (2) infrastructure and (3) education. The most often mentioned suggestion (16.7 of students) was that the government should in some way take up the full responsibilityTable 4. Students' suggestions for how to improve birth registration and tie to possession of own certificate, Kwale, Kenya. total respondents a What should government do to raise birth registration? Register automatically instead of parents. Improve education about registration Enforce the law. Assist parents (non financial). Increase hospital births. Improve access or ease of registration. Reduce cost. Don't know. ASP015K biological activity Explain order FPS-ZM1 importance for school/exams. Explain importance (general). Ask third party to talk to them. fpsyg.2017.00209 Explain importance for ID/citizenship. Take action: do the registration for them. Explain importance for getting job. Explain importance for following law. Don’t know. 16.7 15.2 12.7 11.6 8.7 7.8 5.3 7.4 32.1 20.3 9.7 8.7 7.8 4.9 1.7 9.5 36.7 50.0 62.7 38.2 65.9 43.2 32.0 34.3 40.1 58.3 56.5 29.3 51.4 30.4 37.5 42.2 18.1 (8, N = 473)* 21.2 (8, N = 473)** who possess certificate fpsyg.2017.00209 X2 Answer x cert. possession (df, N) bHow would you encourage your parents to register your birth?Notes: All respondents were elementary school students in grades six to eight.a bPercentages account for individuals who did not answer specific questions. Answer rate for all questions was > 97 . * and ** indicate statistical significance at the .05 and .01 levels, respectively.Comparisons assessed via Chi square. First number in parentheses refers to degrees of freedom. Second number indicates total sample size. doi:10.1371/journal.pone.0149925.tPLOS ONE | DOI:10.1371/journal.pone.0149925 March 3,12 /How Would Children Register Their Own Births?from parents and therefore complete the process automatically while minimizing parent involvement. This typically involved the suggestion that the government dispatch some official representative (chief or other National level official) to visit children or parents in order to complete the registration process. Foll.Ere found to be significant in the model (indicated by asterisks accompanying the Wald statistics). The decision was made to use a logistic regression and Odds Ratios (ORs), rather than prevalence rates or relative risk (RR), following suggestion [32, 33] that in non-controlled sampling cases where there is high outcome prevalence (above 10 ), ORs are a more appropriate summary measure with a more tenable assumption of homogeneity across all individuals in a sample population. Note therefore that the estimates should not be interpreted as approximated risk. The model (p < .001, 2 [24, N = 473] = 165.97, Nagelkerke R2 = 40 , correctly predicted cases = 74.2 ) indicated that only three factors--grade level, school, and whether or not one's sibling had a certificate--remained significant predictors. Odds ratios indicated that students in grade eight had significantly higher odds of possessing a birth certificate when compared to lower grade levels, while having a sibling with a certificate was associated with higher likelihood of the student him/herself possessing a certificate.Students' suggestions for improving certification of their own birthsStudents' own ideas for improving certification rates are reported in Table 4. Regarding how the government could improve registration, student answers--which were given in a free style with no prearranged answer choices--showed a rather even distribution between the themes of (1) direct action, (2) infrastructure and (3) education. The most often mentioned suggestion (16.7 of students) was that the government should in some way take up the full responsibilityTable 4. Students' suggestions for how to improve birth registration and tie to possession of own certificate, Kwale, Kenya. total respondents a What should government do to raise birth registration? Register automatically instead of parents. Improve education about registration Enforce the law. Assist parents (non financial). Increase hospital births. Improve access or ease of registration. Reduce cost. Don't know. Explain importance for school/exams. Explain importance (general). Ask third party to talk to them. fpsyg.2017.00209 Explain importance for ID/citizenship. Take action: do the registration for them. Explain importance for getting job. Explain importance for following law. Don’t know. 16.7 15.2 12.7 11.6 8.7 7.8 5.3 7.4 32.1 20.3 9.7 8.7 7.8 4.9 1.7 9.5 36.7 50.0 62.7 38.2 65.9 43.2 32.0 34.3 40.1 58.3 56.5 29.3 51.4 30.4 37.5 42.2 18.1 (8, N = 473)* 21.2 (8, N = 473)** who possess certificate fpsyg.2017.00209 X2 Answer x cert. possession (df, N) bHow would you encourage your parents to register your birth?Notes: All respondents were elementary school students in grades six to eight.a bPercentages account for individuals who did not answer specific questions. Answer rate for all questions was > 97 . * and ** indicate statistical significance at the .05 and .01 levels, respectively.Comparisons assessed via Chi square. First number in parentheses refers to degrees of freedom. Second number indicates total sample size. doi:10.1371/journal.pone.0149925.tPLOS ONE | DOI:10.1371/journal.pone.0149925 March 3,12 /How Would Children Register Their Own Births?from parents and therefore complete the process automatically while minimizing parent involvement. This typically involved the suggestion that the government dispatch some official representative (chief or other National level official) to visit children or parents in order to complete the registration process. Foll.
