Spss for windows ver 24

In calculating the sample amplitude of this study, the power (power of the test) was determined to be 0.80; the type 1 error was determined to be 0.05 for each variable. The descriptive statistics for the continuous (quantitative) variables are expressed as medians, averages, standard deviations, and minimum and maximum values. Whether the continuous variables in the study were distributed normally or not were examined using the Kolmogorov–Smirnov test (N >50). Nonparametric tests were conducted because the variables were not distributed in a normal fashion. The Mann–Whitney U test was employed to compare the measurements for each pregnancy group. Receiver-operating characteristic analysis was employed to determine the cutoff values of the parameters of FSH and ADM in the pregnancy groups from the standpoint of their values of sensitivity and specificity. Spearman’s correlation coefficients were calculated for the purpose of determining the relationship among variables. A Chi-square test was used to examine the relationship between the categorical variables. Statistical significance level was accepted as %5 in calculations. Finally, the statistical package for the social sciences (SPSS; IBM SPSS for Windows, Ver. 24) was employed to conduct the calculations.

The data were statistically processed using the IBM SPSS for Windows, ver. 24.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics (percentages, means, and standard deviations) were done for general characteristics (age, spouse age, education level, occupation, religion, monthly income, body mass index, stage, type of treatment, exercise activity, number of exercise, and exercise time), and the usability responses of experts and users.

The normality of the distribution was examined using the Shapiro-Wilk test and Kolmogorov-Smirnov test and they showed that the distribution of numerical variables deviated statistically significantly from normal distribution. Skewness and Kurtosis values also showed deviation from normal distribution. Because of that, the median (Mdn) and interquartile range (IQR) were calculated and reported. As the distribution of numerical variables deviated statistically significantly from normal distribution, the non-parametrical Mann-Whitney U test and the Kruskal Wallis test were conducted to test differences in the variables of interest and the relationships were assessed via the Spearman’s correlation coefficient (rho). Univariate linear regression analysis was performed to evaluate the predictive properties of independent variables. A probability level of p ≤ 0.05 was considered statistically significant. All statistical analyses were carried out using the SPSS for Windows, ver. 24.0 (IBM Corp., Armonk, NY, USA).

수집된 자료는 IBM SPSS for Windows ver. 24.0과 AMOS ver. 24.0 (IBM Corp., Armonk, NY, USA)을 이용하여 분석하였다. SPSS 프로그램을 이용하여 기술 통계, 차이검정, 상관관계, 신뢰도 분석을 하고, AMOS 프로그램을 이용하여 확인적 요인 분석과 경로 분석을 시행하여 변수의 타당도, 모형적합도, 변수 간 직접효과, 간접효과, 총 효과, 설명력 등 구조방정식모형을 검증하였다.
측정변수 간 다중공선성을 확인하기 위하여 상관계수, 공차(tolerance), 분산팽창계수(variance inflation factor, VIF)로 분석하였다. 구조방정식의 기본 가정인 정규성을 검증하기 위하여 일변량 정규성 검증은 왜도와 첨도로, 다변량 정규성 검증은 다변량 첨도지수로 검토하였다. 다변량 정규성을 가정하는 최대우도법을 이용하여 분석하였다. 확인적 요인 분석은 표준화된 요인 부하량(standardized factor loading; standardized estimates), 임계치의 유의성(critical ration [C.R.]>1.965), 개념신뢰도(construct reliability), 평균분산추출(average variance extracted, AVE)을 통해 타당도 검증을 하였다.
가설모형의 적합도는 절대적합지수인 χ², χ²/degree of freedom (df), goodness of fit index (GFI), standardized root mean square residual (SRMR), root mean square error of approximation (RMSEA), 증분적합지수인 Tucker-Lewis index (TLI), comparative fit index (CFI), incremental fit index (IFI), 간명적합지수인 parsimonious normed fit index (PNFI), Akaike information criterion (AIC) 등을 이용하여 평가하였다. 경로 분석은 경로계수의 유의성을 통해 검증하였고, 부트스트래핑(bootstrapping) (500회)을 적용하여 직접, 간접, 총 효과의 유의성을 확인하였다.