Term
|
Definition
(Secondary Data)
Inferring something about a whole group of objects from knowledge of a few members of that group |
|
|
Term
|
Definition
| Inferring a conclusion from some premises. E.G. Boyscouts wear khaki, therefore a boy in khaki is a boyscout. |
|
|
Term
|
Definition
| information collected and analyzed without being influenced |
|
|
Term
|
Definition
| information based on data collected systematically. The scientific method consists of the collection of data through observation and experimentation, and the formulation and testing of hypotheses. Empirical refers to the use of working hypotheses that are testable using observation or experiment. |
|
|
Term
|
Definition
| Defining general rules pertaining to subjects' behavior in order to merge existing information, identify new information and predict future events |
|
|
Term
|
Definition
| a set of interrelated propositions that offer an explanation about relationships between 2 or more variables. It's provisional. |
|
|
Term
|
Definition
| Operationally defined concepts that can take on more than one value. The dependent variable is affected by the independent variable. |
|
|
Term
|
Definition
| The causal value in a relationship |
|
|
Term
|
Definition
| the passive variable in the relationship |
|
|
Term
|
Definition
| Research conducted for the purpose of advancing knowledge with little concern for any immediate outcomes. It's driven by a scientist's curiosity or interest in a scientific question. The main motivation is to expand man's knowledge, not to create or invent something. Basic research is also called academic research, because we always look at the literature review; uses research hypotheses. |
|
|
Term
|
Definition
(Primary Research)
Research designed with a practical outcome in mind, under the assumption that some group will derive special benefits. It is designed to solve practical problems of the modern world, rather than to acquire knowledge for knowledge's sake. One might say that the goal of the applied scientist is to improve the human condition. In applied research, there is no literature review available, so we may ask questions in order to specify the problem. |
|
|
Term
| Characteristics of Good Research |
|
Definition
Relevant
on time (up-to-date)
efficient (min input for max output)
accurate
ethical |
|
|
Term
|
Definition
| refers to moral values or principles generally governing the conduct of an individual or group |
|
|
Term
| Researchers' Duties Towards Respondents |
|
Definition
Tell the truth
no privacy invasion
avoid emotional abuse
consider respondents' convenience |
|
|
Term
| Researchers' duties towards sponsors |
|
Definition
avoid unnecessary research
avoid dealing with irrelevant research issues
avoid wrong data analysis
avoid conducting studies with no competences or skills
no overcharging
no violation of confidentiality
avoid conflicts of interest |
|
|
Term
|
Definition
|
|
Term
|
Definition
Common in basic research --> rare in applied research.
It indicates the current knowledge in the area and it's based on academic literature. Thanks to the literature review, we can avoid duplication and can develop/refine the research hypothesis.
E.G. scientific magazines, textbooks, journals |
|
|
Term
|
Definition
Research Questions (or Research Hypotheses)
Research boundaries
All research objectives |
|
|
Term
|
Definition
(Applied Research due lack of literary review)
Asking questions in order to specify the problem.
Ad effectiveness |
|
|
Term
|
Definition
(Basic Research)
Formulated by stating a relationship between a dependent and independent variables. |
|
|
Term
|
Definition
| the issues the research won't deal with |
|
|
Term
|
Definition
-Used to structure the research
-to show how all the major parts of the research project work together to try to address the central research questions.
It's the glue that holds all of the elements in a research project together.
TYPES
1.Exploratory
2. Descriptive
3. Causal (experimental research) |
|
|
Term
|
Definition
| type of research conducted because a problem hasn't been clearly defined. it often relies on qualitative data collection methods (in-depth interviews and focus groups) |
|
|
Term
|
Definition
| it often relies on quantitative data collection methods (surveys) or on qualitative data collection methods (observation) |
|
|
Term
|
Definition
| Explores the effect of one variable on another (experimental research) |
|
|
Term
| Quantitative Data Collection |
|
Definition
(Surveys)-- more accurate than Qualitative
phone
mail
door-to-door
mail intercept
internet |
|
|
Term
| Comparing Quantitative Data Collections Methods |
|
Definition
Flexibility
most: face-to-face
least: internet and mail
Cost
highest: face-to-face
lowest: internet and mail
Speed
fastest: phone
slowest: internet and mail
Response Rate
highest: phone
lowest: mail
Respondents' patience
highest: mail and internet |
|
|
Term
| Qualitative Data Collection Methods |
|
Definition
Used for:
-definition of research problem
-formation of hypothese
-generation of new ideas
-investigation of sensitive issues
-construction of questionnaires
Problem: less accurate than quantitative data collection methods because use unrepresentative samples.
Examples: in-depth interviews
focus groups
observations |
|
|
Term
|
Definition
(Basic Research)
Stats not gathered for the immediate study, but for some other purpose.
It saves the research time and money, but is less accurate, relevant, and up to date than primary data.
E.G. municipalities, internet, libraries, academic literature |
|
|
Term
|
Definition
(Applied Research)
Information collected specifically for the purpose of the investigation. We don't collect primary data if there is viable secondary info available. |
|
|
Term
|
Definition
a method of obtaining objective knowledge about the world through empirical observations.
A fundamental requirement of the scientific method is that all hypotheses and theories must be tested against observations of the natural world, rather than resting solely on intuition. Hence, science is considered to be methodologically empirical in nature. |
|
|
Term
|
Definition
| information collected by a third party for business purposes |
|
|
Term
| Primary Data Collection Methods |
|
Definition
1. observation
2. surveys
3. causal research
4. exploratory research |
|
|
Term
|
Definition
The researcher obeserves people in their natural environment without interfering.
Problem: there may be an ethical problem, and the major disadvantage is that you get only what you see. |
|
|
Term
|
Definition
| based on certain objective criteria |
|
|
Term
|
Definition
| based on subjective opinion of the observer. In unstructured obsercation the observer attempts to describe all the relevant behavior that occurs. This is done by keeping a mental record of what happens that is later developed into a full description of the behavior. |
|
|
Term
| Participating observation |
|
Definition
| a set of research strategies which aim to gain a close and intimate familiarity with a given group of individuals and their practices through an intensive involvement with people in their natural environment. |
|
|
Term
|
Definition
| Pros: large quantity of gathered info and versatility |
|
|
Term
|
Definition
homogeneous groups of 6-12 people that discuss a certain topic or product in a comfortable atmosphere. it's perceived as a stimulating, dynamic, interactive, spontaneous, creative, and quick process, due to:
The Snowball Effect
Hands-on type of experience.
CONS:
-not a representative sample
-individual may take over the group
-situation is artificial
-analysis is difficult and not clear |
|
|
Term
| Characteristics of a Good Moderator |
|
Definition
1. Assertive kindness.
2. Takes initiatives while being calm
3. tolerant
4. flexible
5. sensitive
6. open-minded
He must integrate all participants into the discussion and take their feelings into consideration |
|
|
Term
|
Definition
| a scientific process in which the researcher manipulates one of more independent variables and observes the changes in the dependent variable. |
|
|
Term
|
Definition
variables other than the manipulated variable that affect the response of the test units and hence the results of the experiment
Types:
1. History
2. Maturity
3. Testing Effect |
|
|
Term
|
Definition
| The applicability of experimental results to situations external to the experimental context. Is it applicable in the real world? |
|
|
Term
|
Definition
| all possible things that might have happened between the 2 times of observation. the longer the time span between the observations the higher the risk of history |
|
|
Term
|
Definition
| effect due to the fact that people are constantly changing and becoming more mature. The longer the time span between observation the higher the risk of maturity |
|
|
Term
|
Definition
| occurs when people act differently because they know they're being observed (similar to Hawthorne Effect) |
|
|
Term
|
Definition
the researcher compares the effect of different kinds of manipulations on the same experimental group.
E.G. price study |
|
|
Term
|
Definition
| the researcher wants to test whether 2 or more independent variables influence a dependent variable and explore the interaction between them |
|
|
Term
|
Definition
organized in an isolated artificial environment.
1. Internal Validity = HIGH
2. External Validity = LOW
3. Cost= LOW
4. Time low |
|
|
Term
|
Definition
1. Internal Validity = LOW
2. External Validity = HIGH
3. Cost = HIGH
4. Time = LONG |
|
|
Term
| Quasi- Experimental Design |
|
Definition
offers the researcher some degree of control, but there is no random assignment of subjects, as there is no true experimental design.
-empirical approach--> lack of random assignment
E.G. tracking studies (service improvement) |
|
|
Term
| Internal Validity in Taste Tests |
|
Definition
-Random order of tasting
-rest between each tasting
-personal evaluation (no group influences)
-hiding brand names
-control intervening factors (e.g. smoking)
-usage patterns
-noise odor free venue
-avoid saturation
- feeling of full effect (surpassed max. utility) |
|
|
Term
|
Definition
the assignment of number by rules to objects in order to gain understanding.
Phenomena -> Numbers -> Understanding
Measurement= True Score + Systematic Error + Random Error |
|
|
Term
|
Definition
-Error related to validity
-easily corrected |
|
|
Term
|
Definition
-error related to consistency (more critical)
-can't do any comparison |
|
|
Term
|
Definition
mental stage that people have in a certain environment.
3 Components
1. Behavioral Knowledge
2. Behavioral Affect
3. Behavioral Intention |
|
|
Term
|
Definition
Each Number is Different from the other but it's value has no particular meaning
Numbers assigned to attributes of objects for the purpose of identifying these objects
(e.g. 1= color, 2=price, 3= size etc.) |
|
|
Term
|
Definition
Each number is either bigger or smaller than the other (ranking)--generally used in social sciences.
PROBLEM: can't measure the average, so must use QUASI-INTERVAL sCALE
e.g. unlikely, neutral, likely etc.
It's |
|
|
Term
|
Definition
| the more sophisticated the sample, the more we can use large point scale |
|
|
Term
|
Definition
| same number of positive and negative answers plus one neutral answer |
|
|
Term
|
Definition
| each number has a title. The more sophisticated the sample, the less we use itemized scales |
|
|
Term
|
Definition
1. Multiple Choice
2. Open-ended
3. Dichotomous (e.g. yes/no)
4. order (ranking)
5. Likert scale (quas-interval) |
|
|
Term
|
Definition
-use simple and unambiguous words
-avoid leading questions
-avoid hidden assumptions
-avoid estimates
-avoid double-barred questions ( 2 Q's in 1) |
|
|
Term
|
Definition
| we open a questionnaire with easy and interesting questions and we end with complex, sensitive, uninteresting questions (e.g. demographics). |
|
|
Term
|
Definition
-use when the population size is small
-when want to get accurate results
-want to get info pertaining to small subgroups of the population |
|
|
Term
|
Definition
-when the population is large
-when facing budget and timing constraints |
|
|
Term
|
Definition
| group of people that share some characteristics and from which data can be gathered and analyzed |
|
|
Term
|
Definition
| choosing some units of the population |
|
|
Term
|
Definition
any type of element that makes up a sample.
e.g. people, households |
|
|
Term
|
Definition
a list of the entire population that's used to create a random sample.
e.g. telephone book |
|
|
Term
|
Definition
1. Define the relevant population according to the Research Objectives
2. Define Sample Frame
3. Choose sampling method
4. Determine sample size
5. Execute Sampling |
|
|
Term
|
Definition
a sampling method where the probability of any sampling unti to be included is known (greater than 0 but not necessarily known).
- more accurate
- have some kind of control
4 TYPES:
1. Simple Random Sampling
2. Systematic Sampling
3. Stratified Sampling
4. Cluster Sampling |
|
|
Term
| Non-Probability Sampling Methods |
|
Definition
a sampling method where the probability of any sampling unit to be included is unknown
4 TYPES:
1. Judgmental Sampling
2. Snowball Sampling
3. Convenience Sampling
4 Quota Sampling |
|
|
Term
|
Definition
(Non-Probability Sampling Method)
Selection of items based on the judgement of an individual.
It can't be used for statistical purposes |
|
|
Term
|
Definition
(Non-Probability Sampling Method)
Technique for developing a research sample where existing study subjects recruit future subjects from among their acquaintances. (Friend brings a friend)
The sample group starts small, but appears to grow like a snowball. As the sample builds up, we gain enough data to use for the research. This sampling technique is often used in unique, hidden population, which are difficult for researchers to access, such as drug users.
*In general, there no sampling frame* |
|
|
Term
|
Definition
(Non-Probability Sampling Method)
the method of choosing items arbitrarily and in an unstructured manner from the sampling frame.
The researcher find it easier to go to a particular part of the population we want to investigate and when we have no other choice.
*often used in Exploratory Research-- a problem not yet defined* |
|
|
Term
|
Definition
(Non-Probability Sampling Method)
The population is first segmented into mutually exclusive subgroups and then judgement is used to select the subjects or units from each segment based on a specified proportion.
-Selection of sample is non-random, therefore may be BIASED because not everyone gets a chance of being selected |
|
|
Term
|
Definition
(Probability Sampling Method)
Each individual is chosen randomly and entirely by chance, such that each individual has the same probability of being chosen at any stage during the sampling process
Each subset of 'k' individuals has the same probability of being chosen as any other subset of 'k' individuals.
It's possible that the sample won't be completely random |
|
|
Term
|
Definition
(Probability Sampling Method)
Similar to simple random sampling, but we use a systematic method to choose the sample.
E.G. Every 10th name of the telephone book.
*Number of Intervals: K=N/n |
|
|
Term
|
Definition
(Probability Sampling Method)
When the population consists of a number of distinct and homogeneous categories, the frame can be divided into separate strata. A sample is then selected from each "stratum" separately, creating a stratified sample. WE use this method to ensure that particular groups within the population are adequately represented in the sample and to improve efficiency by gaining greater control on the composition of the sample. |
|
|
Term
|
Definition
(Probability Sampling Method)
Sometimes it's cheaper in time and money to cluster the sample in some way.
E.G. by selecting respondents from certain areas only, or certain time-periods only.
*Each cluster has the same chance of being selected |
|
|
Term
|
Definition
1. Budget Constraints
2. Expected Number of groups in the sample
3. Population Variance
4. Accuracy
5. the "magic number" (n=200) |
|
|
Term
|
Definition
| it's important to have at least 100 units in each group |
|
|
Term
|
Definition
The greater the variance, the great the sample size should be
*direct relationship between population variance and sample size* |
|
|
Term
|
Definition
the greater the sample size
- the smaller the sample error
- the greater the accuracy |
|
|
Term
| Sample Error for Large Samples |
|
Definition
|
|
Term
| Sample Error for Small Numbers |
|
Definition
| (N <, 10,000): (N-n)/(N-1)*(1/√n) |
|
|
Term
|
Definition
| In General, when n=200, we have a good sample |
|
|
Term
|
Definition
(Non-Sampling Errors, Non-Observation)
part of the population the researcher didn't cover |
|
|
Term
|
Definition
(Non-Sampling Errors, Non-Observation)
In survey sampling, many individuals who are identified as part of the sample may be unwilling to participate or impossible to contact.
In this case, there is a risk of differences between the willing and unwilling, leading to SELECTION BIAS in conclusions.
A survey's quality is represented by its Response Rate, and not by the number of people that did answer.
The sample should be representative of the entire population and can't only include extreme cases of satisfaction/ dissatisfaction |
|
|
Term
| Solutions for Low Response Rate |
|
Definition
-Callbacks
-Quota sampling |
|
|
Term
|
Definition
| The degree to which an instrument can predict future behavior |
|
|
Term
|
Definition
| When it's obvious that what was measured actually represents what was supposed to be measured. Face validity is highest in in-depth interviews |
|
|
Term
|
Definition
| Measuring a phenomenon simultaneously using 2 instruments knowing that one instrument is valid |
|
|
Term
|
Definition
| Does the instrument include all the necessary components to measure a phenomenon? |
|
|
Term
|
Definition
| What the instrument is in fact measuring |
|
|
Term
|
Definition
| Measures the level consistency of an instrument |
|
|
Term
|
Definition
| The degree to which the research instrument is able to identify true differences among respondents |
|
|
