Schwarzenberg T L, Loreti P, Sgricia S, Acconcia P, Tiberi R, Buffone M R, Coltellacci M
Istituto di Puericultura, Università degli Studi di Roma, La Sapienza.
Minerva Pediatr. 1997 Apr;49(4):129-37.
In this paper we present growth height standards obtained by an auxological study on a primary school population of central Italy (Lazio), in the 1991-1992 and 1992-1993 school-years.
4,175 children were included in the study: 2,024 females and 2,151 males, aged from 6 to 11 years. The sampling rate corresponds to more than 98% of the total primary school-population of the areas mentioned. The statistical analysis of the collected data was carried out by the Dipartment of Statistics of the University of Rome, using the widely available statistics packages STAT-GRAPHICS and SAS.
The analysis of height mean and standard deviation showed an increasing trend of height related to age, without significant differences by sex. The mean heights for females and males were practically the same in each age class. Height variance increases with age, too. This fact necessitates the use of the weighted regression technique to evaluate the functions that express the height growth trend. The relationship between height and age is linearlike and almost coincides for the two sexes. The height percentiles were calculated and the relative curves were developed, based on the distribution of normalized residuals. The centile curves confirm that there are no significant differences in the growth trends for both sexes and therefore the lines are overlapping. Substantial differences exist only in the end centiles (3th and 97th) and age extremes (6 and 11 years), possibly because of the small number of subjects.
在本文中，我们展示了通过对意大利中部（拉齐奥）一所小学1991 - 1992学年和1992 - 1993学年学生进行人体测量学研究获得的身高增长标准。
该研究纳入了4175名儿童，其中2024名女性和2151名男性，年龄在6至11岁之间。抽样率超过上述地区小学总人口的98%。罗马大学统计系使用广泛可用的统计软件包STAT - GRAPHICS和SAS对收集的数据进行了统计分析。
身高均值和标准差分析表明，身高随年龄呈上升趋势，男女之间无显著差异。每个年龄组中女性和男性的平均身高几乎相同。身高方差也随年龄增加。这一事实使得有必要使用加权回归技术来评估表示身高增长趋势的函数。身高与年龄的关系呈线性，且两性几乎重合。基于标准化残差的分布计算了身高百分位数并绘制了相关曲线。百分位数曲线证实两性的生长趋势无显著差异，因此曲线相互重叠。仅在极端百分位数（第3和第97）以及年龄极值（6岁和11岁）处存在实质性差异，可能是由于样本数量较少。
