Question Details

Assignment
1. Propose a sampling technique you would use for your study, and briefly describe the method of recruitment.  Also specify how you would collect the data.
2. Specify the dependent and independent variables for the study and how would you operationalize these variables. Then describe these variables identifying them as continuous or discrete. Do you know what values these variables take in the general population or in other studies?
Please upload this SLP component at the end of Module 3.
SLP Assignment Expectations:
Develop and submit a 2-3 page paper in which you present your study's sampling plan, including your sampling frame and sampling procedures; recruitment strategy; and independent and dependent variables, their characteristics (nominal/ordinal/interval/ratio AND discrete versus continuous), how you plan to operationalize each, and the values these typically "take on."
)
Overview
In this module we will discuss the fifth step in the research process: sampling, data collection, and types of data.
 
The Research Process
____________________
Formulate a Question
↓
Review the Literature
↓
Develop Hypotheses
↓
Design the Study
↓
Collect the Data
↓
Analyze the Data
↓
Interpret the Results
↓
Write the Report
 
 Sampling Definitions
Sample: a number of individual cases that are drawn from a larger population
Study Population or Sampling Frame: the group of sampling units or elements from which a sample is actually selected; the list from which a sample is selected.
Probability Sample: a sample that gives every member of the population a known (nonzero) chance of inclusion.
Non-probability Sample: a sample that has been drawn in a way that doesn’t give every member of the population a known chance of being selected.
Probability samples are generally more representative of the populations from which they are drawn as compared with non-probability samples.
Probability Sampling
Simple Random Sampling: The most common type of probability sampling.  Each member of the population has an equal and independent chance of being selected to be part of the sample.
Steps in simple random sampling: 
•         Define the population from which you want to select the sample
•         List all the members of the population
•         Assign numbers to each member of the population
•         Establish criterion to select the sample you want and use random number tables
Systematic Sampling: involves selecting every kth element from a list of population elements, after the first element has been randomly selected
Selection interval (k) = population size / sample size
Steps in systematic sampling:
•         Determine the selection interval
•         Choose one observation on the list at random
•         Once the starting point is determined use the selection interval to select your sample
Systematic sampling is easier than random sampling, but can introduce bias if the sampling frame is cyclical in nature.
Stratified Sampling: A procedure that involves dividing the population into groups or strata or sub-groups defined by the presence of certain characteristics and then random sampling from each of the strata
Stratified Sampling can be done proportionately and disproportionately.
Cluster Sampling: A procedure that involves randomly selecting clusters of elements from a population and subsequently selecting every element in each cluster for inclusion in the sample
Multistage Sampling: A procedure that involves several stages, such as randomly selecting clusters from a population, then randomly selecting elements from each of the clusters
Non-probability Sampling
Quota Sampling: A non-probability sampling procedure that begins with a description of the target population (e.g., its proportion of males, females, of people of different age groups, education, etc.)
•         It does not randomly select from the population a subset of all the elements
•         Usually done on a first-come first included basis.
•         Sampling stops when enough are included in each category
Purposive Samples: A non-probability sampling procedure that involves selecting elements based on the researcher- judgment about which elements will facilitate his or her investigation
Used for many exploratory studies and qualitative research. It differs from quota sampling by restricting the sample population to a very specific population and then using all of the subjects available
Snowball Sampling: A sampling procedure that involves using members of the group of interest to identify other members of the group; can be used when population listing are unavailable.
Convenience Sampling: This procedure involves selecting participants that are readily accessible to the researcher. Examples: those in attendance at a meeting/class; interviewing people in a mall; first 200 patients admitted to a medical unit.
Cheap, but potential for bias related to what motivated people to volunteer and segment of the population that is missed because they were not available. Click here to read about recruitment of participants.
 
Data Collection
Methods of data collection are chosen based on the research question, availability of resources, feasibility, and literature review (what methods were used by other researchers). Data collection methods have to be reliable and valid (see Module 2).
Types of data collection
Physiological, Anthropometrical, and Biochemical Measures. Examples: blood pressure, cholesterol levels, measured body weight
Methods of physical examination should be designed to reduce variation within and between observers. Time must be spent with the data collectors to ensure that they are all using the same techniques. Standardization of  laboratory assays can be improved by careful specification of the method of collection and storage and by rigorous quality control of the analysis.
Related concept: Inter-rater variability: variability in the results of data collection between data collectors.
Despite all precautions, observer differences may persist. Observers should be allocated to subjects in a random way and examiner's identity should be entered on the record.
Preexisting Records: The information desired has already been obtained; cuts down time on the data collection process. The data on the records must be obtained in similar fashion and recorded in such a way that consistency can be obtained.
Interviews: May be structured with a specific set of questions or unstructured with just a general theme to start the interview. Interviews can be taped - or the researcher may take field notes as the interview progresses
Questionnaires: May be either self administered (that is, completed by the subject) or administered at interview. Self administered questionnaires are easier to standardize because the possibility of systematic differences in interviewing technique is avoided. On the other hand, they are limited by the need to be unambiguously understood by all subjects. Can use open ended or closed ended items. Closed ended questions are more readily answered and classified, even though they can miss details. Questions should be designed with data coding and data entry in mind.
Recommendations for questionnaire design:
•	The language used should be clear and simple.
•	Two short questions, each covering one point, are better than one longer question which covers two points at once.
•	The order of questions should take into account the sensitivities of the 
person to whom they are addressed
•	Questions should be designed to facilitate recall. As a check on the reliability of information, it may sometimes be helpful to include overlapping questions.
•	Important note: responses to questions should encompass all possibilities, so that subject can ALWAYS answer the question.
Click here for more on questionnaires and interviews.
A pilot survey before starting the main study is recommended.
Variables (Data)
Variables are measurable characteristics of people, objects, or events; the information we are describing and analyzing. The aggregate of our observations comprise our data set.
Independent Variables: studied for their potential or expected influence
Dependent Variables: the outcome, or influenced variables
Operationalization: Specific manner in which one measures or manipulates variables in a study; defining variables so as to make them measurable. Click here to learn more about operationalization of study variables.

Types of Variables
Nominal (Discrete, Categorical):
Variables for which the set of all possible values fall into a finite set of mutually exclusive and exhaustive classes. The values of nominal variables need not be numerically meaningful: addition, subtraction, multiplication, and division do not necessarily make sense.Examples: sex (male, female), color (red, yellow, blue etc), exposed/not exposed, with heart disease/without heart disease, marital status (single, married, widowed, divorced)
Variables with just 2 categories are also called dichotomous.
Ordinal or "Rank":
A nominal variable whose classes or categories have a natural, logical, order. 
Examples: quality (poor, fair, good, excellent), academic level (freshman, sophomore, junior, senior, graduate), frequency of behavior (never, rarely, often, very often), order of finish in an election, respiratory distress (absent, mild, moderate, severe).
Interval and Ratio (Continuous Variables):
Interval Variables. For these variables all possible values are numbers, and subtraction makes sense (intervals are meaningful). Example: temperature.
Ratio. All possible values are numbers, and multiplication and division make sense (ratios are meaningful), i.e., zero (0) means "none". Examples: height, weight, number of children, blood pressure, grams of food
Continuous variables can be transformed and analyzed as categorical variables by establishing cutoffs between ranges of values. For example height in inches can b