Induction of Psychological Experiment Design and SPSS Statistical Analysis in Graduation Design Project

Graduation design projects every year are inseparable from user research methods. Psychological experiment methods are important empirical research methods. The soul of the experiment is the effective control of variables. The purpose of experimental control is to explore the reasons for the occurrence, development and changes of human psychology and behavior. Variables are properties of something that are variable in quantity or quality. Variables include task variables (stimulus variables), environmental variables, and body variables. In a specific experiment, variables are divided into independent variables, dependent variables, and irrelevant variables. Here we need to talk about irrelevant variables in particular. Those relevant variables that the experimenter does not use for research are called additional relevant variables, or simply irrelevant variables, also known as control variables. Irrelevant variables in the experiment must be controlled, and methods such as elimination method, constant method, balance method, offset balance method, randomization method, statistical control method, main experimenter and subject control, etc. can be used. For details, please refer to a Q&A on Zhihu: “心理学行为实验中，怎么控制无关变量？在实验设计和统计上分别该如何处理？“

Experimental design is mainly a plan to control experimental conditions and arrange experimental procedures. Its purpose is to find out the relationship between experimental conditions and experimental results, make correct conclusions, and test hypotheses to solve problems.

Depending on whether the same subjects are used in various independent variables and various treatment levels, experimental design can be divided into three types: within-subjects design, between-subjects design and mixed design. Within-subjects design is the treatment of all situations in which each subject must accept the independent variable. It is divided into three sub-types: pre- and post-experimental design, timed series design and design to offset experimental conditions. A between-subjects design assigns the same number of subjects to different levels of the independent variable or different independent variables. In order to overcome the shortcomings of between-subjects design, randomized group design and matched group design can be used. A mixed design refers to a study in which some independent variables are arranged according to a within-subjects design and some independent variables are arranged according to a between-subjects design.

According to the strictness of experimental control conditions, experimental design can be divided into three types: true experimental design, quasi-experimental design and non-experimental design. The more commonly used types here include completely randomized design, completely randomized factorial design, randomized block design, Latin square design, repeated measures design, crossover design, nested design and split-plot design. For information on the use of the above experimental design methods and their corresponding SPSS statistical analysis methods, please refer to the literature as follows.

“丁国盛. 李涛编著. SPSS统计教程——从研究设计到数据分析. 北京: 机械工业出版社, 2014”.

参考译文

毕业设计课题中的心理学实验设计与SPSS统计分析归纳

关键词：毕业设计，心理学实验，SPSS

每年的毕业设计课题都离不开用户研究方法，心理学实验方法就是重要的实证研究方法。实验的灵魂是对变量的有效控制，实验控制的目的是探索人的心理和行为发生、发展和变化的原因。

变量是指在数量上或质量上可变的事物的属性。变量包括作业变量（刺激变量）、环境变量、机体变量，在一个具体的实验中，变量分为自变量、因变量和无关变量。这里需要特别说一下无关变量，实验者不用于研究的那些相关变量称为额外相关变量，或简称为无关变量，也被称为控制变量。实验中的无关变量必须加以控制，可以采用排除法、恒定法、平衡法、抵消平衡法、随机化法、统计控制法、主试和被试控制等方法。详细内容可参考知乎上的一篇问答：“心理学行为实验中，怎么控制无关变量？在实验设计和统计上分别该如何处理？”

实验设计主要是控制实验条件和安排实验程序的计划，其目的在于找出实验条件和实验结果之间的关系，做出正确的结论，来检验解决问题的假设。

根据在各种自变量及各种处理水平中是否用相同被试，可将实验设计分为被试内设计、被试间设计和混合设计三种。被试内设计是每个被试须接受自变量的所有情况的处理，分为实验前后设计、定时系列设计和抵消实验条件的设计三种子类型。被试间设计就是把数目相同的被试分配到自变量的不同水平或不同的自变量上。为了克服被试间设计的缺点，可采用随机组设计和匹配组设计。混合设计是指在一个研究中有些自变量按被试内设计安排，有些自变量按被试间设计安排。 根据对实验控制条件的严密程度的不同，可将实验设计分为真实验设计、准实验设计和非实验设计三种。这里较为常用的有完全随机化设计、完全随机析因设计、随机化区组设计、拉丁方设计、重复测量设计、交叉设计、嵌套设计与裂区设计等类型。有关上述实验设计方法及其对应的SPSS统计分析方法的使用可参考文献“丁国盛. 李涛编著. SPSS统计教程——从研究设计到数据分析. 北京: 机械工业出版社, 2014”。