Research Writing课堂笔记

Cole 发表于 2008-01-06 13:15:10

Research Writing课程中文名称是：研究写作，
主要讲授学术界（国际通用准则哦）最新的研究方法和数据运用及统计常识。
临考复习，顺便把课堂笔记整理出来吧，方便以后参考~
btw, anyone who wants a Chinese version of the following notes has to wait~cause it takes time to get the translation done, esp. at the end of the semester;)

Research Method

Is a statement or saying true or false? how can we find out?

we research! we mainly apply quantitative method during the research.

Remember: research those problems which are solvable and answerable!

Research is a process, it takes time!

Research is a displine which means it is a whole system that we need to learn!

Research is of scientific nature which means it is rules governed!

How do we know we know & how do we know what we know?

1. Tradition---tells many things, and people seldom challange its truth.

2. Tenacity ---what is right or true for once may not be true for ever.

3. Intuition e.g."I have no reason to spend money with!" then something happens!

4. Illogical reasoning e.g long fishs taste better than short ones.

5. Authority---get the right authority and don't go to a right authority for wrong knowledge! don't ask someone those things they never know!

6. Scientific research---it is empirical! it produces reliable knowledge!

plus: 1---5 provide unreliable knowledge.
research is to test theory and produce knowledge.

What do we research? 1. Difference. 2. Relationship.

By what means can we run the research? T-test (a pair of data, T/F test)

ANOVA/ MANOVA (couples of data)

Limitations of Sciences:
1. science can not answer all our questions
2. science can only answer empirical questions (based on sensory data)

Casual relationship---it is one way, it cannot be reversed! Cause--->Effect

It can only be test or proved through experiments.

Correlation---it is reciprocal causation (互为因果), or it is caused by other external causes.

it is absolute correlation.

Positive correlation 正相关

Negative correlation 负相关

Pearson Correlation 皮尔逊相关（双变量）

Correlation coefficient 相关系数

What are Data? 1. Data are facts; 2. Data are countable, measurable and observable.

Categories of Data:

1. Categorical data 名类/类别变量

数据特点：a.两组数据是相互排斥的

b.数据是穷尽的

c.数据间不能进行算术上的运算

此类数据既不能作相关计算，也不能进行差异比较。

2. Ranking / Ordinal data 序数变量

数据特点：a.两组数据是相互排斥的

b.数据是穷尽的

c. 数据间不能进行算术上的运算

d.可以对数据进行排序

此类数据不能做差异对比。（差异对比主要就是计算数据的平均数）

3. Interval data 区间变量

数据特点：a.数据（范围）是连续的

b.能够进行算术运算

此类数据可以做差异对比或相关计算。

4. Ratio data 比率变量

与区间变量在本质上十分接近，只是在比率变量中 “零” 不等于“无”。

以上四种类型的变量，序数越小，级别越高——高级变量可以变为低级变量，而低级变量不能变为高级变量！

Pretest / Piolet : 3-5 min

皮尔逊相关系数: (Product moment Pearson Correlation)

r描述的是两个变量间线性相关强弱的程度。r的取值在-1与+1之间，若r>0，表明两个变量是正相关，即一个变量的值越大，另一个变量的值也会越大；若r<0，表明两个变量是负相关，即一个变量的值越大，另一个变量的值反而越小。r的绝对值越大表明相关性越强，但这里并不存在因果关系！若r=0，表明两个变量间不是线性相关，但可能是其他方式的相关（如曲线方式）。

Willing to Communicate-WTC, by McCrskey 在本族语言环境下

Kassing—Intercultural WTC-IWTC

Communication anxiety-CA

Communication apprehension

* p<0.05

** p<0.01

*** p<0.001

P stands for possibility—不存在的可能性。要带有星号才存在相关关系。

如果结果没有*，则p=100=>测试项之间不存在关系！

二次数据分析：

在分析数据前一定要先做data clean! 而且一定要保存一份原始数据！！RAW DATA!

已经清理好的数据要另名存盘，以防数据混淆！

Sig—significant

N.—样板数

Linear Regression线性回归

Multiple Regression 多元回归（数据超过两组）

MODE—模数，在数据中出现频率最高的

Median—中位数—数列中数—比MEAN值要稳定，可以避免数列受最大值或最小值影响。

Research procedures:

1. select a problem

2. review existing research and theory—LR

3. develop hypothesis or research questions

4. determine an appropriate methodology/research design

5. collect relevant data

6. analyze and interpret the results

7. present the results in an appropriate form.

Academic research—主要做理论研究

Private sector research—主要为解决问题

Why research?

1. solve problems—applied research

2. resolve disputes—applied research

3. find truth—exploratory research

4. develop theory—academic research

applied—solve real life problems.

academic—theory generating

Process:

1. it’s an on-going process

2. steps are not that rigid

3. researchers need to adjust their methods during the process.

Problem Identification:

1. specific

2. doable of answerable questions

3. interesting

4. not trivial

Find a Research Method:
1. How can I collect the data?
2. What method will I use for the research?
3. Measurements design (questionaire)
4. Selecting a sample.

Selecting a Research Topic. Simplest approach is the most efficient.

Literary Review: (Justification)
Qs to ask--what has been done in this area?
                what has been found in previous study?
                what suggestions do other researchers make for further study?
                what has not been investigated?
                how can the proposed study add to our knowledge of the area?
                what research methods were used in previous studies?

Hypothesis?------theory generating.
A hypothesis is a formal statement regarding the relationship between variables and is tested directly. The predicted relationship is either true of false.

Research Questiong:
A research question is a formally stated question intended to provide indications about something; it is not limited to investigating relationships betweem variables. It is generally used when a researcher is unsure about the nature of the problem under investigation. Testable hypotheses are often developed from information gathered during the research question phase of a study.

What's a Theory?
1. a guess about ways things are
2. an idea about how something works
3. how things relate to each other
4. what happens when something happens

Theory---good or not?---->simple, serve/inspire research(heuristic)

Research Design.

Questionnaire Design---for survey
Topics appropriate for survey: Survey must be used for descriptive, explanatory and exploratory purposes.
(tips for questionnaire design--efficient use of space, faster to complete, comparability of responses.)

Questionnaire Design: Concept, Variable, Measurement and Scale.
Concept: an abstract and unobservable object (what to ask in a survey)
Variable: the empirical, concrete, observable version of a concept (how to ask a question)
Measurement: the assignment of numerals to a variable (how to design answers)
Level of measurement: nominal scale, ordinal scale, interval scale, ratio scale.
Scale: special format of measurement
Common scales in survey: Likert Scales, Semantic Differential Scales.

Guidlines for asking questions:
# choose appropriate question forms
# make items clear
# avoid double-barreled questions
# respondents must be competent to answer
# respondents must be willing to answer
# questions should be relevant
# short items are best
# avoid negative items
# avoid biased items and terms
# no value judgment questions

How to judge the quality of a survey question?
1. Validity: the question invites honest, accurate answers.
2. Reliability: the question invokes consistent answers if asked in different time or location.
3. Pre-test is an effective means to ensure the quality of a questionnaire.

Question wording:
1. use clear and unambiguous terms
2. KISS principle
3. offer a neutral option in answers to a multiple-choice question or a Likert Scale.

Data Analysis:
# Quanlification of Data 定量
# Univariate Analysis 单变量分析
# Bivariate Analysis 双变量分析
# Multivariate Analysis 多变量分析

Chi-Square: 卡方检验（只适合于名列变量！）
Test 1. a distribution for GOODNESS OF FIT.
       2. two variables for independence.
      3. proportions for homogeneity.

O--observed frequency 实际频数
E---expect frequency   平均值（推测出的数值）
d.f.---自由度=number of catagory-1 （样板数减一）

将运算结果对表，结果〉表中指数——无关联
                           结果〈表中指数——有关联
Steps for Chi-Square calculation:
1. forming hypothesis
2. d.f.=degree of freedom
3. 查表，determine the Apha value
4. find the critical value
5. caltulate the expect frequency
6. calculate the Chi-square value
7. decission : 对表查---接受H(A) [有偏好]或者接受H(o)[无偏好]

Cross Table：(双向卡方) or contingency table.

Read Excel Data into SPSS: (Excel 4.0 or upper)
1. go to "filt/open"
2. look inthe directory for the Excel file
3. change "file of type" to "Excel(*.xls)"
4. select the file
5. Important: select "read variable names" before clicking OK!!

SPSS 的使用方法暂时没办法细述，而且本期测试不需要用SPSS，呵呵，今天我就总结到这里~~

课程中Sherwin还穿插讲了APA格式的运用和论文（按国际标准！）写作的一些内容，
有很多繁琐的事项，不能一一总结——APA到目前已经更新到第五版了~细目还在不断更改
=.=想象一下吧~~

研究写作用APA的格式比较多，文学分析的就用MLA的多~另外还有一些Chicago Style。。。格式的东西，五花八门，难以一统！

plus, 我每次都是先写论文后弄格式的~呼呼，懒得去记那些条条框框~

（据说在国外有专门帮别人弄论文格式的，一份论文，大约300p，200美金——帮你按照APA, MLA或者其他什么麻烦的格式弄好~嘻嘻）

关键词(Tag): 研究方法课堂笔记

阅读199次评论个人主页扔小纸条文件夹: 读书讲座

收藏: QQ书签 del.icio.us 订阅: Google 抓虾

发表评论

* 昵称
	已经注册过? 请登录新用户请先注册以便能显示头像及追踪评论回复
Email
网址
* 评论
表情
* 验证码	请输入您看到的数字 (注册登录就不用每次输入讨厌的验证码了)