第 2 章 像研究者一樣思考

學習目標 讀完本章你應該了解： 1. 採用科學思考的專業研究者使用之專業術語。 2. 制定一個好的研究假說需要哪些條件。 1. 採用科學思考的專業研究者使用之專業術語。 2. 制定一個好的研究假說需要哪些條件。 3. 合理推論的需求，強化了企業研究的成果。

研究與直覺 「我們千萬別忽略了經驗與智慧這兩件事，那是神奇靈感、直覺的來源。我與你有更多共同的經驗，就更可能碰到重複的活動與處境，也能協助我向你學到更多；我愈有智慧，就愈能提煉這些經驗，找到其中的關聯與模式。」 Jeffrey Bradshow，資料庫搜尋軟體的創造者

研究的語言 研究使用的 專有名詞 變數Variables 模型 Models 理論 Theory 構念Constructs 操作性定義Operational definitions 命題／假說Propositions/ Hypotheses 概念性架構Conceptual schemes 概念Concepts Several terms are used by researchers to converse about applied and theoretical business problems. A concept is a bundle of meanings or characteristics associated with certain concrete, unambiguous events, objects, conditions, or situations. The importance of conceptualization is discussed in the following slide. A construct is a definition specifically invented to represent an abstract phenomena for a given research project. Exhibit 3-1, a depiction of job redesign constructs, is provided in Slide 2-13. A conceptual scheme is the interrelationship between concepts and constructs. An operational definition defines a variable in terms of specific measurement and testing criteria. An example of an operational definition is provided in Slide 2-14. A variable is used as a synonym for the construct being studied. Slides 2-15 through 2-20 expand on different types of variables. A proposition is a statement about observable phenomena that may be judged as true or false. (Slide 2-21) A hypothesis is a proposition formulated for empirical testing. (Slides 2-22 through 2-25) A theory is a set of systematically interrelated concepts, definitions, and propositions that are advanced to explain or predict phenomena. Slide 2-26 shows an example of a theory. A model is a representation of a system constructed to study some aspect of that system. Slide 2-27 shows an example of a model.

研究的語言 成功的研究 清晰表達概念 共同理解的概念 We must attempt to measure concepts in a clear manner that others can understand. If concepts are not clearly conceptualized and measured, we will receive confusing answers.

重新設計工作中的概念與構念 Exhibit 3-1 Exhibit 3-1 illustrates some of the concepts and constructs relevant to job redesign. The concepts at the bottom of the exhibit (format accuracy, manuscript errors, and keyboarding speed) are the most concrete and easily measured. Keyboarding speed is one just concept in the group that defines a construct that the human resource analyst calls Presentation Quality. It is not directly observable like keyboarding speed. It is a term used to communicate (a label) the combination of meanings presented by the three concepts. Concepts at the next level are vocabulary, syntax, and spelling. As they are related, the analyst groups them into a construct she calls language skill. Language skills is placed at a higher level of abstraction in the exhibit because two of the concepts that comprise it, vocabulary and syntax, are more difficult to observe and measure. The construct of job interest is not yet measured nor are its components specified. Researchers often refer to such constructs as hypothetical constructs because they are inferred only from the data—they are presumed to exist but no measure tests whether such constructs actually exist. If research shows the concepts and constructs in this example to be interrelated, and if the connections can be supported, then the analyst has the beginning of a conceptual scheme. One exercise you can try is to have students attempt to identify the concepts/constructs in the hypothetical construct…job interest, and discuss which elements are truly measurable…and how.

操作性定義 我們如何定義變數「學生就讀的年級」？ 新鮮人 低於30學分 二年級學生 30至59學分 三年級學生 60至89學分 新鮮人 低於30學分 二年級學生 30至59學分 三年級學生 60至89學分 四年級學生 90學分以上 Operational definitions are definitions stated in terms of specific criteria for testing or measurement. The specifications must be so clear that any competent person using them would classify the objects in the same way. If a study of college students required classifying students by class level, a definition of each category would be necessary. Students could be grouped by class level based on self-report, number of years in school, or number of credit hours completed. Credit hours is the most precise measure.

變數具備可供研究的特質 變數 事件 行動 特性 特質 屬性 In practice, the term variable is used as a synonym for the property being studied. In this context, a variable is a symbol of an event, act, characteristic, trait, or attribute that can be measured and to which we assign categorical values. The different types of variables are presented on the following slides. 屬性

變數的類型 二分法 Dichotomous 不連續 Discrete 連續 Continuous 男／女 僱用／非僱用 種族背景 教育程度 宗教團體 連續 Continuous 收入 溫度 年齡 For the purposes of data entry and analysis, we assign numerical values to a variable based on that variable’s properties. Dichotomous variables have only two values that reflect the absence or presence of a property. Variables also take on values representing added categories such as demographic variables. All such variables are said to be discrete since only certain values are possible. Continuous variables take on values within a given range or, in some cases, an infinite set.

獨立變數與依變數的同義詞 獨立變數 Independent Variable (IV) 預測變數 Predictor 推測原因的變數 Presumed cause 刺激變數 Stimulus 預測自…的變數 Predicted from… 先行變數 Antecedent 操控變數 Manipulated 依變數 Dependent Variable (DV) 準則變數 Criterion 推測影響的變數 Presumed effect 反應變數 Response 預測成…的變數 Predicted to…. 結果變數 Consequence 測量產出變數 Measured outcome Exhibit 3-2 Exhibit 3-2 presents the commonly used synonyms for independent and dependent variables. An independent variable is the variable manipulated by the researcher to cause an effect on the dependent variable. The dependent variable is the variable expected to be affected by the manipulation of an independent variable.

圖表2-3 各類變數之間的關係(a)

圖表2-3 各類變數之間的關係(b)

圖表2-3 各類變數之間的關係(c)

干擾變數 Moderating Variables (MV) 特別是對較年輕的員工（MV）來說，導入每週工作四天的制度（IV），會提高生產力（DV）。 喪失了採礦工作（IV）， 導致接受高風險以賺取家計所需收入的行為（DV），尤其對教育程度低的當地居民（MV）而言，更是如此。 Moderating variables are variables that are believed to have a significant contributory or contingent effect on the originally stated IV-DV relationship. Whether a variable is treated as an independent or as a moderating variable depends on the hypothesis. Examples of moderating variables are shown in the slide.

外生變數Extraneous Variables (EV) 對新顧客類型（EV-控制）而言，由薪資報酬變成傭金制度（IV），將導致每位員工的銷售生產力（DV）增加，尤其較年輕的員工（MV）更是如此。 學歷較低的當地居民（EV-控制） ，一旦喪失採礦工作（IV），就會導致接受高風險工作的行為（DV） ，特別是臨近演習區（MV），更會發生這種狀況。 Extraneous variables are variables that could conceivably affect a given relationship. Some can be treated as independent or moderating variables or assumed or excluded from the study. If an extraneous variable might confound the study, the extraneous variable may be introduced as a control variable to help interpret the relationship between variables. Examples are given in the slide.

中介變數Intervening Variables (IVV) 透過提高工作滿意度（IVV），由薪資報酬變成傭金制度（IV），將帶來更佳的銷售生產力（DV）。 特別是在提供免費贈品（MV）之際，促銷活動（IV）會增加存款（DV），但主要是影響小額存款人（EV-控制），由於提高存款動機（IVV），才會造成這結果。 An intervening variable (IVV) is a factor that affects the observed phenomenon but cannot be measured or manipulated. It is a conceptual mechanism through which the IV and MV might affect the DV.

命題與假說 Propositions and Hypotheses 品牌經理Jones（個案）的成就動機（變數）高於平均值。 Z公司的品牌經理（個案）的成就動機（變數）高於平均值。 一般化 A proposition is a statement about observable phenomena that may be judged as true or false. A hypothesis is a proposition formulated for empirical testing. A case is the entity or thing the hypothesis talks about. When the hypothesis is based on more than one case, it would be a generalization. Examples are provided in the slide.

假說的形式 描述性 Descriptive 假說 在底特律城（個案），我們的馬鈴薯薄片市場占有率（變數）穩居13.7%。 某些美國城市（個案）正遇到預算的難題（變數）。 研究問題 在底特律城，我們的馬鈴薯薄片市場占有率是多少？ 某些美國城市正遇到預算的難題嗎？ A descriptive hypothesis is a statement about the existence, size, form, or distribution of a variable. Researchers often use a research question rather than a descriptive hypothesis. Examples are provided in the slide. Either format is acceptable, but the descriptive hypothesis has three advantages over the research question. Descriptive hypotheses encourage researchers to crystallize their thinking about the likely relationships. Descriptive hypotheses encourage researchers to think about the implications of a supported or rejected finding. Descriptive hypotheses are useful for testing statistical significance.

關係性假說 Relational Hypotheses 關聯性 Correlational 比起35歲（含）以上的女性，年輕女性（35歲以下）較少購買我們的產品。 套裝銷售的數量和景氣循環的水準同步變化。 因果性 Causal 家戶所得（IV）增加，會導致所得儲蓄（DV）百分比增加。 對特定雜貨店的忠誠度（IV）提高，會增加購買由這商店贊助私有品牌貨品（DV）之可能性。 A relational hypothesis is a statement about the relationship between two variables with respect to some case. Relational hypotheses may be correlational or explanatory (causal). A correlational hypothesis is a statement indicating that variables occur together in some specified manner without implying that one causes the other. A causal hypothesis is a statement that describes a relationship between two variables in which one variable leads to a specified effect on the other variable.

假說的角色 指導研究的方向 辨別相關的事實 建議最適合的研究設計 提供組成研究結論的架構 This slide presents the functions served by hypotheses.

良好假說的特性 配合研究目的 良好的假說是 可檢定的 比其他假說更好 The conditions for developing a strong hypothesis are more fully developed in Exhibit 3-4. 比其他假說更好

研究中的理論 Exhibit 3-5 What is the difference between theories and hypotheses? Theories tend to be complex, abstract, and involve multiple variables. Hypotheses tend to be simple, limited-variable statements involving concrete instances. A theory is a set of systematically interrelated concepts, definitions, and propositions that are advanced to explain or predict phenomena. To the degree that our theories are sound and fit the situation, we are successful in our explanations and predictions. The product life cycle, shown in Exhibit 3-5, is an example of a theory.

推理的角色 Exhibit 3-7: Business models are developed through the use of inductive and deductive reasoning. As illustrated in Exhibit 3-7, a business model may originate from empirical observations about market behavior based on researched facts and relationships among variables. Inductive reasoning allows the modeler to draw conclusions from the facts or evidence in planning the dynamics of the model. The modeler may also use existing theory, managerial experience or judgment, or facts.

研究中的模型 Exhibit 3-6 A model is a representation of a system constructed to study some aspect of that system or the system as a whole. Models versus Theories a model’s role is to represent or describe A theory’s role is to explain. Models in business research may be descriptive, predictive, and normative. Descriptive models are used for complex systems because they allow for the visualization of numerous variables and relationships. Predictive models forecast future events and facilitate business planning. Normative models are used for control, because they indicate necessary actions. Exhibit 3-6, shown in the slide, is a distribution network model called a maximum flow model used in management science. In this example, a European manufacturer of automobiles needs an increased flow of shipping to its Los Angeles distribution center to meet demand. However the primary distribution channel is saturated and alternatives must be sought. Models allow researchers to specify hypotheses that characterize present or future conditions: the effect of advertising on consumer awareness or intention to purchase, brand switching behavior, an employee training program, or other aspects of business.

科學方法 直接觀察現象 清楚定義變數 清楚定義方法 可驗證、可檢定的假說 排除對立假說的能力 統計證明 自我修正的過程 Good business research is based on sound reasoning because reasoning is essential for producing scientific results. This slide introduces the scientific method and its essential tenets. The scientific method guides our approach to problem-solving. An important term in the list is empirical. Empirical testing denotes observations and propositions based on sensory experiences and/or derived from such experience by methods of inductive logic, including mathematics and statistics. Researchers using this approach attempt to describe, explain, and make predictions by relying on information gained through observation. The scientific method is described as a puzzle-solving activity.

研究者 遇到問題 陳述問題 擬定假說 演繹假說的產出 明確敘述對立假說 設計並處理可驗證的檢定 寫出結論 調整原始的問題 The steps followed by business researchers to approach a problem are presented in the slide.

好奇心，是成為好的研究者之必要條件 Synovate公司的廣告強化一項特性：好奇心，是成為好的研究者之必要條件。 This ad from Synovate reinforces the notion that researchers must be curious. Students can see the Synovate website at www.synovate.com.

合理的推理 講述的方式 揭示Exposition 辯論Argument 演繹Deduction 歸納Induction Exposition consists of statements that describe without attempting to explain. Argument allows us to explain, interpret, defend, challenge, and explore meaning. There are two types of argument: deduction and induction. Deduction is a form of reasoning in which the conclusion must necessarily follow from the premises given. The next slide provides an example of a deductive argument. Induction is a form of reasoning that draws a conclusion from one or more particular facts or pieces of evidence. Slide 2-8 illustrates an inductive argument. 演繹Deduction 歸納Induction

演繹式的推理 市內家庭訪問特別困難且費用昂貴。（前提1） 這次調查涉及大量的市內家庭訪問。 （前提2）   This slide provides an example of a deductive argument. 這次調查訪問將會特別困難且費用昂貴。 （結論） © 2002 McGraw-Hill Companies, Inc., McGraw-Hill/Irwin

歸納式推理 為什麼在促銷活動期間內銷售額沒有增加？ 在促銷期間內，該地區的零售商沒有足夠的存貨滿足顧客的需求。 我們卡車公司的員工正好罷工，以致貨品無法及時運達，降低了促銷的效率。 在促銷期間內颶風來襲，使該地區所有零售店歇業10天。 This slide provides an example of an inductive argument.

銷售額為什麼沒有增加？ Exhibit 3-8 Induction and deduction can be used together in research reasoning. Induction occurs when we observe a fact and ask, “Why is this?” In answer to this question, we advance a tentative explanation or hypothesis. The hypothesis is plausible if it explains the event or condition (fact) that prompted the question. Deduction is the process by which we test whether the hypothesis is capable of explaining the fact. Exhibit 3-8 illustrates this process.

Tracy 的績效 Exhibit 3-9

關鍵詞 辯論（argument） 個案（case） 概念（concept） 概念性架構（conceptual scheme） 構念（construct） 演繹法（deduction） 實證主義論（empiricism） 揭示（exposition） 假說（hypothesis） 關聯性假說（correlational hypothesis） 描述性假說（descriptive 解釋性（explanatory）又稱因果 性假說（causal hypothesis） 關係性假說（relational 假說性構念（hypothetical construct） 歸納法（induction） 模型（model） 操作性定義（operational definition） 命題（proposition） 合理的推理（sound reasoning） 理論（theory） 變數（variable） 控制變數（control variable） 混淆變數（confounding variable, CFV） 依變數（dependent variable, DV）又 稱準則變數（criterion variable） 外生變數（extraneous variable, EV） 獨立變數（independent variable, IV） 又稱預測變數（predictor variable） 中介變數（intervening variable, IVV） 干擾變數（moderating variable, MV）