biology reading site image no navigation top banner image
spacer gif

Understanding methods


Here is the methods section of the article from Obesity (2006: 14, 5, 855–862) entitled: Energy Intake, Diet Composition, Energy Expenditure, and Body Fatness of Adolescents in Northern Greece by M. Hassapidou, E. Fotiadou, E. Maglara, and S. Papadopoulou. Notice the staging of the elements and that some stages are shorter than others.

Click on the elements in the left hand column to see or hide an example of each element or group of elements.

Methods element Example

Sampling technique

A total of 512 subjects, 11 to 14 years of age, in the first, second, and third grade of high school were selected to participate in the study. The city of Thessaloniki was divided into four sections (northern, eastern, western, and southern). One public school from each section was chosen randomly. All pupils from the first three grades of these schools participated in the study. In terms of socioeconomic status, pupils were from all types of families (working class, middle class, and upper class families). Therefore, the sample was representative of the pupil population in Thessaloniki. Ethical approval was given by the Greek Ministry of Education. Two pupils dropped out during the study and eight pupils did not respond properly. The final sample was 502 pupils.
Four groups of adolescents were identified for the study: boys and girls with BMI corresponding to ≥25 kg/m2 at 18 years, classified as overweight, and boys and girls with BMI corresponding to <25 kg/m2 at 18 years, classified as non-overweight, according to the international standards suggested by Cole et al. (19). The final sample consisted of 83 overweight and obese boys, 185 non-overweight boys, 48 overweight and obese girls, and 186 non-overweight girls.

Procedure and Measurements (not statistical)

Anthropometric measurements were taken for all participants. One person conducted the anthropometric measurements to avoid interexaminer variability. Body weight was measured with a digital scale with an accuracy of ±100 grams (seca 707; seca Corp., Hamburg, Germany). Height was measured to the nearest 0.5 cm without shoes with the use of stadiometer (seca 220). BMI was calculated by dividing weight (kilograms) by height squared (meters squared). Skinfold thicknesses at two sites (triceps and subscapular) were measured using a Harpenden skinfold caliper (British Indicator, London, United Kingdom). Skinfolds were measured in duplicate, and the average measurement was used. In case there was a discrepancy of >10% between duplicate measurements, the measurement was repeated. Percentage body fat was estimated from skinfold measurements, using the formula of Slaughter et al. (20).

Food intake was determined by a 3-day weighed dietary diary. One spring balance accurate to ±2 grams was provided to each participant at school. After receiving the appropriate instructions written by two dietitians who participated in the study, the adolescents, with the help of their parents, weighed, described, and wrote down all items of food and drink consumed for 3 consecutive days including 1 weekend day. Food records were obtained after the 3 days of recording. If there were gaps or questions on the food records, dieticians contacted the respondents by telephone.

Twenty-four-hour energy expenditure was calculated by totaling RMR plus energy cost for activity. RMR was calculated using the equation of Schoffield for the specific age (21). These equations have been checked and found applicable for the Greek population in previous studies (3,5). The energy cost for activity has been calculated based on an activity record, kept by the pupils, for 3 consecutive days, on the same days as energy intake. Dietitians explained how to complete the activity records, regarding type and level of exercise during the day. For each participant, RMR value was multiplied with an average metabolic value according to his/her activities, using the tables by Ainsworth et al. (22)

Data analysis

The diaries completed by the adolescents have been coded using McCance and Widdowson's food tables (23). All data were analyzed using the Microdiet food program (version 8; University of Salford, Salford, UK). Greek food recipes were added in the basic database according to the Food Composition Tables and Composition of Greek Cooked Food and Dishes (24).

Statistical analysis

Data are expressed as means ±standard deviation, and the statistical differences between overweight and non-overweight adolescents of the same sex were determined with an unpaired Student's t test, after checking for normality of distribution of the dependent variable. Otherwise, a Mann-Whitney test was performed. All data analyses were performed by using the SPSS statistical package (version 8.6; SPSS, Chicago, IL), and the level of statistical significance was set at p < 0.05.

Click Next to proceed to the results section.