A cross-sectional survey of college students was conducted in Hunan Province, China, during a one-month period of March 8 to April 8, 2018.
Sample and setting
The survey was conducted in Hunan Province, China, which is located in the middle of China. In this province, student enrolment in regular institutions of higher education is 22.1 persons per 10,000 population, and there are 109 schools that are regular institutions of higher education . Three comprehensive universities were selected by convenience sampling. Each university has a population of approximately 30,000 students who come from different provinces in China.
Inclusion criteria: 1) Aged between 18 and 30 years old; 2) Full-time college students of regular institutions of higher education; 3) Informed consent and voluntary participation. Exclusion criteria: 1) Non-Chinese student; 2) Pregnant; 3) Have children. Students were recruited by open recruitment as advertised by poster. These posters, including a brief description of the study and a QR code, were posted on the notice boards of each university, canteen and dormitory in the university to be seen by all students and staff. By scanning the QR code using a mobile phone, students could obtain a brief description of the study, an informed consent form and the questionnaire (Questionnaire star software). The questionnaire star software (Ranxing information technology company, Changsha, China), the first and largest domestic online questionnaire survey and test platform, has sent more than 24.6 million questionnaires and collected more than 1.6 billion questionnaires from respondents. The online survey was open from March 8 to April 8, 2018. Participants who completed the questionnaire were provided a reward incentive via the internet.
The sample size was estimated by the formula . Uα/2=1.96，δ=0.05，π=0.569  , put into the formula, the calculated sample size was about 376. In order to avoid errors, the sample could be expanded by 10%~20%, 376~451. In total, 867 individuals completed the online questionnaire during the 1-month collection period, and 799 questionnaires were valid.
Questionnaire and study variables
The questionnaire consisted of 25 items covering 4 domains. It was pilot tested twice with 30 college students to ensure that the instrument was understandable. The pilot data were not included in the analysis.
Participants stated their age in years, gender, place of birth, whether they were of the only child generation (yes/no), their discipline, relationship status, and sexual orientation.
Intentions of childbearing
One question investigated the personal intention to have children (yes/no), and one 0- to 10-point response scale surveyed the personal perceived importance of childbearing .
The Chinese version of the Cardiff Fertility Knowledge Scale (CFKS-C) was translated into Chinese from the original Cardiff Fertility Knowledge Scale (CFKS) developed by Boivin et al. (2013) and shown to have satisfactory validity and reliability. The CFKS consists of 13 items. Three areas of knowledge about fertility are measured: risks for reduced fertility, misconceptions about fertility, and basic facts about infertility. A three-point scale of “true”, “false”, or “do not know” was used to rate all items. One point was given for a correct answer, and zero points were given for an incorrect or unknown answer. Scores were reported as percentages that were equal to points divided by the total number of questions. In the original CFKS, the internal consistency coefficient alpha (Cronbach’s α) was 0.79 . The CFKS-C was developed through Brislin's translation/back-translation method and modified through cultural adaptation and semantic analysis. The Cronbach's α coefficient was 0.827, the test-retest reliability was 0.826, and the Scale-Content Validity Index (S-CVI) was 0.950 with the Item-Content Validity Index (I-CVI) ranging from 0.875 to 1.000 .
Fertility health education
In addition, three questions were used to investigate students’ perceived knowledge of fertility issues: (Ⅰ) their perceived education level of fertility-related knowledge (not at all educated, somewhat educated, educated, highly educated), (Ⅱ) whether or not they discussed fertility issues with family (yes/no), and (Ш) where they gained most of their knowledge about fertility issues (media and internet, schools, family, friends, doctors/gynaecologists, non-government organizations, other).
Descriptive statistics were used to describe the sociodemographic variables and fertility knowledge. Categorical data were compared using Chi-squared tests. T-tests and analysis of variance (ANOVA) were used to compare the total scores on the CFKS-C between sociodemographic categories. Pearson’s correlation was used to examine the association between the importance of childbearing and the CFKS-C score. The data were assessed for errors before double-entry computer input. Statistical significance was defined as a two-sided P value less than 0.05. All analyses were performed using SPSS (version 22.0) software (IBM Corp., Armonk, New York, USA).
Ethical review and approval were performed by the IRB of behavioural and nursing research in the School of Nursing of Central South University (Project Number 2017028). Permission to use the CFKS was obtained from the authors who developed it. A cover letter was presented to respondents to explain the aim and process of the study before the questionnaire was shared. Participation in this study was voluntary, anonymous and confidential. Returning the questionnaire was voluntary and done by pressing the “confirm submit” button via the internet, so no consent form was required. To maintain anonymity, participants were asked not to provide their name or telephone number. Collection of online data complied with the “Code of Conduct of Marketing Research”. All data collected were treated anonymously and confidentially.