UNIT IV

OBJECTIVES AND HYPOTHESES Lesson Objectives The student will be able to: 1. Write research objectives/questions/hypotheses for survey, ex post facto, and experimental research. 2. Evaluate each for appropriate content criteria. 3. Develop a hypothesis. 4. Describe the relationship of theory to hypothesis development. 5. Discuss the testing of hypotheses. 6. Write null and alternative hypotheses for statistical testing. DEVELOPING SPECIFIC OBJECTIVES I. Statement of the problem in a proposal or report of research should be accompanied by a list of specific objectives for the study. Specific objectives take the form of --A. List of specific questions or objectives to be answered, and/or B. List of hypotheses to be tested II. Form that the specific objectives will take is influenced by the "end sought" or purpose of the study Purpose--"End Sought"

Form for Specific Objectives

Description of status (Surveys) Objectives or Questions to be answered; and/or Comparative hypotheses

- How cases distributed on variables - Exploratory relationships between variables

Explanation - Prediction of Outcomes (Relational/Associational studies) - Explanatory relationships between variables that are expected to exist on the basis of some rationale

Research hypotheses to be tested if ex post facto; questions or objectives if correlational

Control Outcomes (Experimental Studies) - Cause-effect relationships between variables

Research hypotheses to be tested

41

______________________________________________________________________________ If type of study is: │ then USE │ Surveys or │ Research Questions Qualitative Studies or │ or Correlational │ Objectives Ex post facto or Experimental

│ │ │

Hypotheses

III. Characteristics of specific objectives -- (criteria for judging) A. Specifically identify the variables being investigated 1. Variables are given operational names -- examples a. instead of saying "achievement" of students, use specific operational variable names like "score on the XYZ achievement test," "GPA at time of graduation," etc. b. instead of "characteristics of students," identify the specific characteristics such as age, gender, grade level, socio-economic status, etc. 2. Indicate what data are to be collected B. If relationships are to be investigated; specific objectives indicate clearly what variables are being related to other variables. C. Indicates the nature (functional or cause-effect) of the relationships being investigated. IV. Developing Hypotheses A. Hypotheses 1. Tentative solution to problem or answer to question 2. Conjectural statement of the relation between two or more variables 3. Specifies the relationships expected between variables B. Criteria for appraising statements of research hypotheses 1. Identify specifically and operationally the variables being related--States a relationship between variables. 2. Identify what variables are being treated as independent and dependent variables. 3. Indicate the nature (type) of relationship expected -- i.e., test cause-effect hypotheses or investigate hypotheses to explain and/or predict outcomes. 4. Specify the direction of the relationship expected. 5. Supported by a rationale--that is, there is some basis for the hypothesized expected relationships.

42

C. Basis (rationale) for hypotheses: Review of Literature 1. Theory – Hypothetical-Deductive Theory; Model/Paradigm a. Systematic ordering of ideas...to be able to inter-relate a set of variables on the basis of the rules of logic b. Provides logical explanations for facts c. A theoretical framework for research 1) sharpens research objectives 2) suggests what variables should be investigated 3) aids in interpreting results 4) makes research cumulative from one study to the next d. Deducing hypotheses from theory -- researcher proceeds in logical fashion from established facts and relationships to the formulation of new hypotheses to be tested empirically. 2. Empirical research -- researcher develops hypotheses based on the findings of related research: like an Inductive Theory 3. Logical argument from expert opinion and/or personal experience – Functional Theory D. Research and Null Hypotheses 1. Research hypotheses -- other names - scientific hypotheses - substantive hypotheses a. These are hypotheses derived from theory and/or related research b. Research hypotheses are stated in the proposal 2. Null hypotheses -- statistical hypotheses a. Hypothesis of no difference or no relationship b. Examples -- (given in class) c. Not stated in the proposal -- Stated in the report of research when data are presented and statistical analysis is made (Chapter 4, Findings/Results). V. Summary for Unit on Developing Specific Objectives - Quote from Krathwohl - p. 28 "Hypotheses should be stated when there is a basis for prediction. State them as succinct predictions of the expected outcomes and findings rather than in the null form. (The latter is an important part of the logic of the statistical test, but it does not belong in the objectives section, and it creates an amateurish impression.) There will be a basis for such predicted outcomes and findings if there is a theoretical underpinning for the study. Questions are most appropriate where the research is exploratory, or where the study is a survey seeking certain facts." NULL HYPOTHESES (Ho) Ho - A null hypothesis is a hypothesis of no difference. It is the condition the researcher holds to be true until evidence shows otherwise. It is done for mathematical convenience, i.e., to apply the statistical test. It says the results of the study could be due to chance. 43

Examples of Ho (1) Ho: µ1 = µ2, or [for comparing means] Ho: µ1 - µ2 = 0 (2) Ho: σ1 = σ2 (3) Ho: ρ = 0

[for standard deviations] [for correlations]

Once Ho is specified, then the alternative hypotheses are listed. These may be listed as H1 or Ha and Ha', or H2. Ha (H1) - Alternatives are stated in two ways: (1) Saying the two groups are unequal (…) H1 or Ha: µ1 … µ2 or

H1 or Ha: µ1 > µ2 H2 or Ha': µ1 < µ2 EX: McGregor and Wilson basketballs have different scuff resistance. [Called a nondirectional or two-tailed test] (2) By predicting the direction (which could go either way)

H1 or Ha: µ1 > µ2 EX: McGregor basketballs are superior to Wilson's. or H1 or Ha: µ1 < µ2 EX: Wilson basketballs are superior to McGregor's. [Called a directional or one-tailed test] Other examples: (I) If you were confident that your new way of teaching remedial reading (1) would improve the reading achievement scores over the old way (2), then your hypotheses would be:

Ho: µ1 = µ2 H1: µ1 > µ2 44

(II) If you were confident that your physical fitness program (1) would reduce the heartbeat rate as compared with non-participants (2) then your hypotheses would be:

Ho: µ1 = µ2 Ha: µ1 < µ2 (III) If you were conducting a workshop to apprize guidance counselors of the roles of school administrators and were not sure if their attitudes toward administrators would improve or decline, then your hypotheses would be:

Ho: µ1 = µ2 H1: µ1 … µ2 or H1: µ1 > µ2 & H2: µ1 < µ2 EXAMPLE OF DERIVING A HYPOTHESIS FROM THEORY Theory Self-perception theory ---People make inferences about their own motives --Whether a person perceives himself/herself to be intrinsically or extrinsically motivated may have important consequences --The test of intrinsic motivation is what a person does when external pressures to engage in the behavior are absent Development of Hypothesis If ... We induce a child to engage in an interesting activity by offering a reward Then ... The child may perceive that activity differently -- rather than being an end itself (intrinsic motivation), the activity may become a means to an end (extrinsic motivation). Because of the reward, the child has a different, ulterior motive for engaging in the activity. So ... (the expected result is) At any other time, when the reward is no longer available, the child may be less likely to engage in the activity on his/her own. Statement of General Hypothesis . . . A person's intrinsic interest in an activity may be decreased by inducing him/her to engage in that activity as an explicit means to some extrinsic goal. Problem Statement Will children who are asked to draw with magic markers in order to obtain an extrinsic reward be subsequently less interested in playing with them than either children who were asked to draw but had not received a reward or children who were asked to draw and had received the same reward unexpectedly? (note: this “statement” was worded as a “question”). 45

Prediction (Deducing the Consequences of the Hypothesis) . . . Hypothesis suggested that an extrinsic reward should undermine later intrinsic interest only when the child expected a reward and agreed to play with the markers as a means of getting it. Specific Research Hypothesis . . . Children who had expected and received an award for their drawing would subsequently play less with (magic) markers in the absence of further rewards, than children who received an award unexpectedly or not at all.

46