research_fundamentals

The collection and analysis of numerical data to describe, explain, predict, or control phenomena of interest.

The collection, analysis, and interpretation of comprehensive narrative and visual data to gain insights into a particular phenomenon of interest.

The formal, systematic application of the scientific method to the study of educational problems.

To decribe, explain, predict, or control educational phenomenon.

A question of interest that can be tested or answered through the collection and analysis of data.

Developing generalizations based on observation of a limited number of related events or experiences.

Arriving at specific conclusions based on general principles, observations, or experiences.

• 1. Identify Problem
• 2. Formulate Hypothesis
• 3. Collect Data
• 4. Analyze Data
• 5. Statement of Conclusions

• 1. Can't answer all questions.
• 2. Can never capture the full richness of the context.
• 3. Measurement error.

Collecting data to determine whether, and to what degree, a relation exists between 2 or more quantifiable variables.

At minimum, information is required about at least 2 variables obtained from a single group of participants.

Attempts to determine the cause, or reason, for existing differences in behavior or status of groups or individuals.

Cause is the grouping variable.

Change or difference is the dependent variable.

• 1. Survey Research
• 2. Correlational Research
• 3. Causal-Comparative Research
• 4. Experimental Research
• 5. Single-Subject Research

Determines and reports the way things are; it involves collecting numerical data to test hypotheses or answer questions about the current status of the subject of study.

• 1. Narrative Research
• 2. Ethnographic Research
• 3. Case Study Research

A Qualitative Research approach to conducting research on a unit of study or bounded system (i.e., individual teacher, classroom, or school as a case).

Also called Ethnography. The study of the cultural patterns and perspectives of participants in their natural settings.

Focus on a site or sites that provide a context to study both the setting and those who inhabit it.

The study of how different humans experience the world around them; it involves a methodology that allows people to tell the stories of their "storied lives".

It is typically focused on a single person.

By Method or Purpose

At least one independent variable is manipulated, other relevant variables are controlled, and the effect on 1 or more dependent variables are observed.

True experimental research provides the strongest results because it provides clear evidence for linking variables.

As a result it offers generalizability, or applicability of findings to settings and contexts different from the one in which they were obtained.

Where experimental researchers compare a single person's behavior before treatment to behavior exhibited during the course of the experiment.

May also study a number of people together as one group, rather than as individuals.

Used to study the behavior change that an individual or group exhibits as a result of some intervention or treatment.

The size of the sample is said to be one.

• 1. Basic and Applied Research
• 2. Evaluation Research
• 3. Research and Development (R & D)
• 4. Action Research

• 1. Simple Random
• 2. Stratified Random
• 3. Cluster Random
• 4. Systematic Random
• 5. Non Random

A quantitative sampling technique deriving a selection using a table of random numbers.

A quantitative random sampling technique where a selection is based on dividing up into subgroups which have roughly equal representation.

A quantitative random sampling technique where intact groups are selected such as a classroom. This is efficient but less representative of a given population.

A quantitative random sampling technique where you take a population and select every Kth individual. For example the population is 100 and you select every 5th person on the list.

• 1. Intensity
• 2. Homogeneous
• 3. Criterion
• 4. Snowball
• 5. Random Purposive

The process of selecting a small number of individuals for a study in such a way that the individuals chosen will be good key informants (collaborators, co-researchers) who will contribute to the researcher's understanding of a given phenomenon.

• 1. Convenience Sampling
• 2. Purposive Sampling
• 3. Quota Sampling

It's the process of including whoever happens to be available in the sample.

Also referred to as accidental sampling and haphazard sampling.

It is the process of selecting a sample that is believed to be representative of a given population. Also referred to as judgement sampling.

The process of selecting a sample based on required, exact numbers, or quotas of individuals or groups of varying characteristics.

A qualitative sampling technique where you select participants who permit study different levels of the research topic.

For example, the researcher might select some good and poor students, experienced and inexperienced teachers, or teachers with small and large classes.

A sample strategy would be to compare the differences of two or more levels of the topic (e.g., good vs. bad students); select 2 groups of about 20 participants from each of the two levels.

A qualitative sampling technique where you select participants who are very similar in experience, perspective, or outlook; this produces a narrow, homogeneous sample and makes data collection and analysis simple.

A sample strategy would be to select a small group of participants who fit a narrow, homogeneous topic; collect data from the chosen participants.

A qualitative sampling technique where you select all cases that meet some set of criteria or have some characteristic; the researcher might pick students who have been held back in two successive years or teachers who left the profession to raise children and then returned to teaching.

A sample strategy is to identify participants who meet the defined criterion; select a group of five or so participants to collect data from.

A qualitative sampling technique where you select a few people who fit a researcher's needs, then using those participants to identify additional participants, and so on, until the researcher has a sufficient number of participants.

Snowballing is most useful when it is difficult to find participants of the type needed.

Sample strategy would be to decide how many participants are needed, let initial participants recruit additional participants that fit the researcher's requirements until the desired number is reached.

A qualitative sampling technique where you select more participants than needed for the study.

For example, if 25 participants were purposively selected by the researcher but only 10 participants could take part in the study, a random sample of 10 from 25 potential participants would be chosen; this strategy adds credibility to the study, although the initial sample is based to purposive selection.

This approach is typically used with small samples.

• 1. Observations: field notes
• 2. Interviews: face to face, phone, focus groups
• 3. Questionnaires
• 4. Archival documents
• 5. Journals
• 6. Audio/video

• 1. Researchable (can be solved by collecting and analyzing data).
• 2. Interesting enough to hold researchers' attention
• 3. Significant (contributes to the field of knowledge)
• 4. Ethical (no harm done).
• 5. Manageable (keep it in scope)

• 1. Solvable by collecting data
• 2. Cannot be answered with a simple yes/no
• 3. States what is the effect of X on Y
• X= independent variable
• Y=dependent variable

• 1. See what has already been done
• 2. Reveals what needs to be done
• 3. Addresses the research questions
• 4. Provides rationale for your hypothesis
• 5. Avoids the mistakes of others
• 6. May suggest ideas not previously considered
• 7. Provides new direction to your research

An educated guess based on theories/other people's research or what you've observed.

Based on sound reasoning consistent with theory or observation.

Provides reasonable explanation for a predicted outcome.

States the expected relationship between variables.

Is testable within a reasonable timeframe.

• 1. Deductive
• 2. Inductive
• 3. Directional
• 4. Non-directional
• 5. Null

Is the degree to which a test consistently measures whatever it is measuring.

Validity assumes reliability.

Is the degree to which a test measures what it is intended to measure.

Validity assumes reliability.

• 1. Unclear test directions
• 2. Ambiguous test items
• 3. Complex vocabulary and sentence structure
• 4. Subjective scoring
• 5. Items on a test not covered in the instruction
• 6. Cheating, either by participants or by someone teaching the correct answers to the specific test items.

• 1. Stability
• -test-retest
• 2. Equivalence
• -2 similar tests
• 3. Internal Consistency
• -requires only one test administered

• 1. Split Half
• 2. Kuder-Richardson
• 3. Chronbachs Alpha

• 1. Achievement tests
• 2. Aptitude tests
• 3. Affective tests
• 4. Projective tests

The effect or outcome

The cause or treatment

A placeholder that can assume any one of a range of values.

• 1. Nominal
• - names categories (i.e. male/female)
• 2. Ordinal
• - ranks from high to low (i.e., class rankings)
• 3. Interval
• - scores on achievement tests (i.e., 90% and 100%)
• 4. Ratio
• - has true zero point (i.e., height, weight, time, distance)

• 1. Likert
• -strongly agree to strongly disagree
• 2. Semantic Differential
• - select a position on a continuum
• 3. Rating Scales
• - various forms (select the statement that best describes you).

• 1. Categorical analysis (qualitative data)
• 2. Statistical analysis (quantitative data)

• 1. Reading/memoing
• 2. Classifying
• 3. Indentifying themes and patterns
• 4. Coding

• 1. Descriptive Statistics
• - describes a set of data
• 2. Inferential Statistics
• - tests of significance between two or more means, allows generalization, significance level

p < .05 or .01

• 1. Frequencies
• -number of times something occurs
• 2. Central Tendency
• -mean, median, mode
• 3. Variability
• -the spread - range, standard deviation, variance
• 4. Relative Position
• -percentile rank, standard scores)
• 5. Relationship
• -correlation coefficient, -1 to 1, Pearson r for ratio and interval, Spearman Rho for ranked data

• 1. Null Hypothesis
• - no difference between variables; any difference will be due to chance.
• 2. Type I error
• - you determine the null hypothesis is false and it is actually true (no difference). False positive (you think you're pregnant and you're not).
• 3. Type II error
• - you determine the null hypothesis is true and it is fase (there IS a difference); False negative (you think you're NOT pregnant and your are). This is less common.
• 4. T-tests
• - test the means of 2 independent samples; significant difference between group A and B.
• 5. ChiSquare
• - used for nominal scales (male/female) compares proportions actually observed to the proportions expected to see if they are significantly different. For example flipping a coin 50 times.
• 6. Anova
• - 2 or more means are statistically different?