Sampling Methods and Sample Populations in Quantitative Mass Communication Research Studies: A 15-Year Census of Six Journals

This study evaluated the sampling methods and sample populations used in all U.S.-based survey and experimental mass communication studies published between 2000 and 2014 in six major journals (N = 1,173). Most studies used nonprobability samples, and more than half used student samples. Experiments used more nonprobability and student samples than surveys. Funded studies used more probability and nonstudent samples than nonfunded studies. Implications of results pertaining to population validity and interpretations of findings for mass communication research are discussed.     

Discovering the Laws of Sampling

That the sample mean and variance are “good” estimates of the corresponding population parameters is easily accepted as “obvious” by students, but the concept of standard error of the mean is often found to be quite a hurdle. That this standard error decreases inversely as the square-root of the sample size, and the mysterious appearance of the Normal distribution, are often taken as magical and incomprehensible effects, and non-mathematical students can often be turned away from further understanding. This article describes a program which provides an experimental framework in which the student can rapidly develop an intuition for the basic properties of sampling.     

The BioSS Challenge - A demonstration of sampling bias

A sampling problem is described which was given to the general public to illustrate the problems of bias and the need for randomisation when sampling.     

Optimal sampling frequency in recording of resistance training exercises

The purpose of this study was to analyse the raw lifting speed collected during four different resistance training exercises to assess the optimal sampling frequency. Eight physically active participants performed sets of Squat Jumps, Countermovement Jumps, Squats and Bench Presses at a maximal lifting speed. A linear encoder was used to measure the instantaneous speed at a 200 Hz sampling rate. Subsequently, the power spectrum of the signal was computed by evaluating its Discrete Fourier Transform. The sampling frequency needed to reconstruct the signals with an error of less than 0.1% was f_99.9 = 11.615 ± 2.680 Hz for the exercise exhibiting the largest bandwidth, with the absolute highest individual value being 17.467 Hz. There was no difference between sets in any of the exercises. Using the closest integer sampling frequency value (25 Hz) yielded a reconstruction of the signal up to 99.975 ± 0.025% of its total in the worst case. In conclusion, a sampling rate of 25 Hz or above is more than adequate to record raw speed data and compute power during resistance training exercises, even under the most extreme circumstances during explosive exercises. Higher sampling frequencies provide no increase in the recording precision and may instead have adverse effects on the overall data quality.     

对linux指定音频文件播放的探讨

文章主要是通过调用ioctl,open,read,write等函数,结合LINUX音频设备文件,实现了对指定音频文件的播放。     

基于ARM的实验室数字电源设计

利用Cortex-M3,在线性电源基础上,结合数字化处理技术实现了一款高精度实验室数字电源。对模拟电源电路进行了改进,使之适合于数字化的处理,并将数字电路引入反馈之中,从而达到很好的数字控制效果,并设计了与计算机上位机的接口,为利用计算机软件控制电源输出作好了进一步扩展的准备。此外,本电源稳定性好,纹波系数高,可扩展性能优良,为实现高精度实验提供了较好的保证。     

基于FPGA的AD转换控制器设计

采用FPGA器件EP1C12实现了对A/D转换芯片AD7822的实时采样控制,并将采集的数据暂存到SRAM中以备后续处理。整个设计在QuartusⅡ环境下,采用Verilog HDL语言描述,给出了硬件电路连接、硬件内部逻辑设计以及测试波形,可用于模拟信号的高速实时采集。     

典型矮塔斜拉桥地震风险概率评估方法研究

鉴于矮塔斜拉桥受力性能的特殊性,引入增量动力分析（IDA）和拉丁超立方体抽样（LHS）评估该结构在地震激励下发生损伤的风险概率。首先采用LHS考虑桥梁构件材料的随机性,与事先选取的地震波相结合形成桥梁-地震动样本集合;其次在确定桥梁有限元模型主要构件损伤指标的基础上采用基于蒙特卡罗（MC）抽样的IDA进行地震易损性分析和地震风险概率评估。为验证方法的正确性和有效性,对一预应力混凝土独塔双索面矮塔斜拉桥结构模型进行地震风险概率评估,结果表明:该方法既充分考虑了材料和地震输入双重随机性又避免了繁琐的积分过程,同时也提高了工程计算的效率及精度。     

脉宽调制控制方法研究

介绍了脉宽调制(PWM)的原理,重点分析了各种正弦脉宽调制(SPWM)方法,其中包括自然采样法,规则采样法等,并利用基于DSP的脉宽调制实验平台对各种SPWM方法进行了比较分析,得出了一些有益的结果。     

大众媒体与共同知识的建构

大众传播不仅具有传播、传承知识的功能,以大众媒体为载体的内容传播本身还是建构知识的过程。本文以此为基本视角,以《北京晚报》和《新京报》关于《北京媒体公信力调查报告》的抽样样本是否具有代表性的争论为案例,解析媒体传播知识的同时也是建构知识的过程,认为知识传播本身可能导致建构某种偏见为知识本身。同时,本文还对描述性研究的抽样原理进行了必要的解释。