DESIGN, VALIDATION, AND RELIABILITY OF SURVEY TO MEASURE KNOWLEDGE OF NUTRITION, WEIGHT CONTROL AND ITS RISKS

The purpose of this study was to design, validate, and test the reliability of an instrument to evaluate knowledge of nutrition, weight control and its risk. The instrument collects information regarding: socio-demographics and athletic status; basic knowledge of nutrition (the diet they follow, nutrients, supplements, energy balance, myths, hydration and habits); and weight control and risks (weight control, eating behaviors, and weight control habits). The design, validation, and testing of the reliability of the questionnaire were done in four phases: a) design and development of the instrument, b) content validation, c) instrument reliability, and d) concurrent validity. The results show that the instrument is suitable for measuring nutrition, weight control and risk knowledge in athletes. The instrument that was developed and validated in this paper can contribute to assessing how the athletes evolve through their different formation stages.


Introduction
Athletes train to achieve success in competition. Ideally, sport practice should involve improving or maintaining the physical, psychological, and social wellbeing of the athletes. However, this is not always true (Sundgot-Borgen, 2002). For example, in sports where the aesthetics or weight control is important (e.g. gymnastics or combat sports), many athletes try to reduce their weight with the theoretical goal of being more successful (Sundgot-Borgen, 2002;Steen, Brownell, 1990). These weight control cycles risk affecting athletes' self-perception and health. Additionally, numerous studies have reported the use of inadequate weight control techniques (Artioli et al., 2010a;Ubeda et al., 2010;Valliant, Emplaincourt, Wenzel, Garner, 2012). There are several possible reasons for the use of these unhealthy procedures in athletes, such as stress, social and environmental factors, or stereotypes.

#0#
Prevention is widely regarded as the key to combating the risks related to weight control (Rust, 2002;Beals, Brey, Gonyou, 1999;Joy et al., 1997). Coaches focus their efforts on increasing their players' knowledge and skill in the sport, but they should also provide information regarding reducing the risks of their athletes being unhealthy. In order to measure their athletes' needs and the effects of specific continuing education about this topic, an instrument to measure their knowledge and habits is needed. The information about the athletes' knowledge or their nutrition and weight control habits will provide useful information for coaches, federations, and managers to ensure healthy habits in their players and emphasize these aspects in the life of athletes. Ideally, monitoring the health of the athletes and the aspects that affect it should be as important as monitoring the physical or technical aspects. In the bibliography review carried out, several instruments that measure the knowledge and habits of athletes with regard to nutrition or weight control risk have been found (Artioli et al., 2010b;Bonci et al., 2008;Brito et al., 2012;Juzwiak, Ancora-Lopez, 2004;Zawila, Steib, Hoogenboom, 2003). However, those instruments were focused on specific aspects such as knowledge of nutrition or nutritional habits. In order to provide useful tools to coaches and researchers to measure how athletes are evolving or the effect of a specific educational programming, a comprehensive instrument is needed. This instrument must provide information from different perspectives or approaches due to the multi-dimensional aspects that are involved. This information should help coaches, dietitians, sport psychologists, athletes' families, etc. in the process of maintaining the athletes' health. The purpose of this study was to design, validate, and test the reliability of an instrument to evaluate athletes' knowledge about nutrition, weight control and its risks.

Method
The design, validation, and testing of the reliability of the questionnaire were done in four stages: a) design and development of the instrument, b) content validation, c) instrument reliability, and d) concurrent validity (Trochim, Donnelly, 2007). The design and development of the instrument involved the use of specific literature about sport nutrition, weight control and its risks. Reviews in the following databases were done: ISI Web of Knowledge, Medline, SPORTDiscus, Google Scholar, EBSCO, and Dialnet. The key word searches included: "nutrition", "knowledge", "weight control", "eating disorders", "athletes", "survey" (as well as their equivalents in Portuguese and Spanish for the Scielo and Dialnet databases, respectively). A review of abstracts was done to select the papers related to the instrument topic. Questionnaires found in the literature (Artioli et al., 2010b;Bonci et al., 2008;Brito et al., 2012;Juzwiak, Ancora-Lopez, 2004;Zawila, Steib, Hoogenboom, 2003), specific literature (Bean, 2010;Bonci, 2009), and literature about creating an original instrument (Hague, Hague, Morgan, 2004;Thomas, 2004) were used as guides. Some of the questions were translated to Spanish and adapted to be used in different sports such as combat sports, gymnastics, or athletes in general. In the process of designing the first draft of the survey, the researchers, a dietician, two coaches of combat sports, two combat sport athletes, and a former gymnast participated in the process of selecting the questions, adapting or wording the questions, and clarifying them.
From this review, a list of key aspects about the female athlete triad was established. Questions were grouped into: a) athletes' characteristics (10 questions), nutrition knowledge (23 questions), nutrition habits (2 frequency charts), knowledge of weight control and its risks (12 questions), habits and perceptions regarding weight control and its risks (19 questions In the second stage, the instrument was sent to four experts in fields related to at least one of the components of the survey, nutrition or weight control. The experts were asked to evaluate qualitative (open questions) and quantitative questions (scale from 1 to 10) from the survey regarding: degree of understanding of the survey's questions; degree of adequacy of the survey's questions, and the need to reduce or include more questions in the survey. The collective suggestions from the experts were considered, and the appropriate changes were made. A descriptive analysis of their answers (i.e. mean, median, and mode) was also done. Following Bulger and Housner (2007), questions with values lower than 7.0 were eliminated, questions with values between 7.1 and 8.0 were modified, and questions with values greater than 8.1 were accepted or accepted with modifications. With the values from the quantitative evaluation done by the experts, the Aiken's V was calculated (Penfield, Giacobbi, 2004).
In the third stage, the reliability of the instrument was calculated. The questionnaire was completed by Spanish wrestlers and rhythmic gymnasts. The four week test-retest procedure was completed by 12 wrestlers (24.16 ±4.87, national level) and 11 gymnasts (16 ±1.16, international level). A final section allowing for comments took into consideration their understanding of the questionnaire, the time taken to complete the survey, and questions or concerns they had with the instrument. Reliability of each item was calculated using the Kappa Index for each of the questions (categorical variables) using the SPSS software.
In the fourth stage, the ability of the instrument to differentiate between athletes of different age groups was measured (Trochim, Donnelly, 2007). Twenty-one under-16 wrestlers and 20 senior wrestlers who participated in the Spanish National Championship were analyzed (2012-2013 season). An inferential analysis of the data (onefactor ANOVA) was done to establish the existence of differences between wrestlers of both ages using the SPSS 21.0 software, with a level of statistical significance set at p < 0.05.

Results
The draft of the survey had 63 questions after the first stage. Sub-scales with 23 points and 12 points were established with regard to the questions pertaining to nutrition knowledge and weight control and its risks, respectively. After reviewing the experts' evaluation of the draft of the survey, vocabulary for four questions were changed. The experts' observations were related to the terms used, the need to clarify the terminology or questions, etc. At this stage, all questions from the draft of the survey had an average score >7.0. The Aiken's V was pertinent (>0.81 for the lowest value). From the score of the test-retest carried out with wrestlers and gymnasts, the total reliability of the questionnaire was calculated (the smallest of these calculations). Intra-class correlation coefficients of 0.615 and 0.609 were found for male wrestlers and female gymnasts, respectively. Regarding the ability of the instrument to differentiate theoretically different age groups of wrestlers (Table 1), significant differences were found in the nutrition knowledge score (p < 0.001) and weight control and its risks score (p<.001) for the wrestlers of different age groups. The final Spanish and English versions of the survey can be found after the references.

Discussion
This paper describes the process done to design and validate a survey to measure knowledge of nutrition, weight control and its risks. In the first stage, the review of the available literature and similar questionnaires was the basis for developing the survey. The information from different sources was translated and adapted to ease understanding and pertinence for athletes from different environments and sports. The combination of internal reviews, mini-pilot studies with athletes and coaches, and the experts' opinions contributed to increasing the clarity, understanding, and proper terminology of the survey. The quantitative evaluation done by experts allowed for the establishment of the pertinence of the sections and questions of the survey (Bulger, Housner, 2007;Escurra, 1989;Padilla, Gómez, Hidalgo, Muñiz, 2007;Zhu, Ennis, Chen, 1998). The levels of content validity found are higher than the proposed minimum (Penfield, Giacobbi, 2004). The level of reliability of the instrument in the test-retest procedure carried out with male wrestlers and female gymnasts was "substantial" (Landis, Koch, 1977). The values showed that it is pertinent at this level (intra-rater reliability). The results show that the survey that was developed has the ability to measure differences between groups that, in theory, are different.
These data show that the instrument can be useful to measure the knowledge of nutrition and weight control and its risks in athletes. The survey has a structure that allows for dividing it into several parts, so it can be used as a whole survey or different parts, depending on the goals and needs. The combination of information related to knowledge and habits allows us to determine what the needs of the athletes are. Future studies are needed about reference values and normative profiles for different sports, genders, and levels of competition in order to properly understand and apply the information provided by the survey.

Conclusions
The process that was followed and the data that were found show that this instrument is suitable for measuring athletes' knowledge of nutrition as well as of weight control and its risks. This instrument can be useful to evaluate the need for specific education and the effect of educational training. Performance sport can involve the risk of developing unhealthy behaviours. Coaches, clubs, institutions, etc. must be proactive to avoid these problems. Prevention is a key aspect to prevent these health problems, and measuring the states and needs of the athletes is part of preventive and proactive actions. More research and specific normative profiles are needed to reduce the risk of inadequate eating and weight control behaviors.

CUESTIONES SOBRE NUTRICION
A continuación se presentan una serie de afirmaciones sobre conocimientos en nutrición. Señala en la columna si estás de acuerdo o no con cada una de ellas.

QUESTIONNAIRE ABOUT NUTRITION, WEIGHT CONTROL AND ITS POSSIBLE RISKS
This questionnaire is part of a research study being carried out at the University of ________. The study is an attempt to assess your knowledge of nutrition, weight control and the possible risks involved with weight control. Since the questionnaire is anonymous, we ask that you respond as sincerely as possible. The data you provide are very important for the knowledge of our sport. To complete the questionnaire, mark an X under the response that best corresponds to your answer, keeping in mind that unless otherwise indicated, there should only be one answer. When necessary, print clearly on the lines provided.

32.
How often do you consume these foods when you are training for your sport (in the current season)? (Mark an X in the corresponding box)

QUESTIONS ABOUT WEIGHT CONTROL AND THE RISKS INVOLVED WITH IT
Next, there are a series of statements about weight control and its risks. Mark in the column whether or not you agree with each statement. Yes No 33. Sudden changes in weight do not affect one's physical or mental health. 34. Increasing training sessions and decreasing intake increases the stress that is put on the body. 35. Skipping meals is justifiable when you need to achieve rapid weight loss. 36. Weight loss causes an imbalance in energy reserves and body fluids. 37. Bulimia nervosa is characterized by large binges followed by self-induced vomiting.
38. An athlete who suffers an eating disorder or who is at risk of suffering one has an increase in performance. 39. The female athlete triad is made up of amenorrhea, osteoporosis, and an eating disorder. 40. Eating disorders can be fatal. 41. The combination of excessive sports practice and a prolonged decrease in food intake can cause bone mineral density loss (osteoporosis). 42. Anorexia nervosa includes the failure to maintain a normal body weight for one's age and height as well as the intense fear of gaining weight. 43. The energetic demands of an athlete vary according to the point in the season. 44. The body needs a minimum level of fat to function correctly (6-13% for men and 12-18% for women). 45. Monitoring your weight is an adequate way to monitor the effect of a change in your diet. 46. When you lose weight, you only lose body fat. 47. A growing person (normally until 18-20 years of age) has different demands than a person who has finished growing. 48. Body composition can change through reduced intake and/or increased exercise. 49. Monitoring urine color is a good method to monitor your level of hydration. 50. When an athlete suffers from an eating disorder, the role of coaches and family is to detect the possible existence of a problem, to refer the athlete to a specialist, and to provide support in the treatment process. Yourself 0 1 2 3 4 5 6 7 8 9 10 Teammates 0 1 2 3 4 5 6 7 8 9 10 Coach 0 1 2 3 4 5 6 7 8 9 10 Physical therapist 0 1 2 3 4 5 6 7 8 9 10 Psychologist 0 1 2 3 4 5 6 7 8 9 10 Parents/family members 0 1 2 3 4 5 6 7 8 9 10 Friends 0 1 2 3 4 5 6 7 8 9 10 Doctor 0 1 2 3 4 5 6 7 8 9 10 Internet/ forum 0 1 2 3 4 5 6 7 8 9 10 Dietician 0 1 2 3 4 5 6 7 8 9 10 Other (_____________) 0 1 2 3 4 5 6 7 8 9 10 56. Have you ever been clinically diagnosed with an eating disorder? (mark all that correspond) ( ) No ( ) Anorexia ( ) Bulimia ( ) Other (which?) __________________ 57. Indicate whether you have ever used the following methods to lose weight. Mark the corresponding box. 58. Indicate the degree of satisfaction you have with your body on a scale from 0 to 10, with 10 being very satisfied _________ 59. Indicate the degree of satisfaction you have with your weight on a scale from 0 to 10, with 10 being very satisfied _________ 60. What is your ideal weight? _________ 61. Do you think you have sufficient knowledge of the following? Respond using a scale from 0 to 10, with 10 being the most knowledge) Nutrition and diet ______ Weight control ______ Risks involving weight control (anorexia, bulimia…..)______ 62. Have you ever received information about the following? (mark any that apply and fill in from whom you received the information) ( ) Nutrition and diet (I've received information from __________________________) ( ) Weight control (I've received information from __________________________) ( ) Weight control risks (anorexia, bulimia…) (I've received information from __________________________) ( ) Other (_________________________) (I've received information from __________________________)

THANK YOU FOR YOUR COLLABORATION!
Comments (use this space to make any observations or comments regarding any aspect touched on in this questionnaire):