CHAPTER - 4 RESEARCH DESIGN * A research design is the arrangement for the collection, measurement and analysis of data * It is a master plan specifying the methods and procedures for collecting and analyzing the needed information * Thus, the research design is the "backbone" of the research protocolCont'd Research design is decisions regarding what, where, when, how much, by what means concerning a research study What is the study about? Why is the study being made? Where will the study be carried out? What type of data is required? Where can the required data be found? * What periods of time will the study include? What will be the sample design? What techniques of data collection will be used? How will the data be analyzed? In what style will the report be prepared?Important Concepts Relating to Research Design 1. Dependant and Independent variables: A variable is anything that varies or changes in value or may assume different numerical or categorical values. In causal-relationship research, the terms dependent variable and independent variable are frequently used.Dependant and Independent variables: Dependent variable * If one variable depends upon or is a consequence of the other variable, it is dependant variable. * It is the "effect" in a cause- and-effect relationship. * Example: In an experiment to test the effect of temperature on plant growth, the plant growth is the dependent variable because its growth is expected to depend on the temperature. Independent variable * isa variable that is expected to influence the dependent variable It is often Cause in a cause- and-effect relationship. Example: In an experiment to test the effect of temperature on plant growth, the temperature is the independent variable.2. Extraneous variables: Variables that are not related to the purpose of the study, but may affect the dependent variable or outcome of study * If not controlled , Extraneous variables could potentially lead to an error. Hence, researcher has to try to identify and control these variables to ensure the validity of the findingsFew Examples * Time of Day: In testing cognitive abilities, performance might differ in the morning versus the evening. If you tested one group in the morning and one in the afternoon, the results may vary. * Temperature: when studying plant growth, unexpected temperature fluctuations can influence results * Participants mood: A participants mood can affect their response in an experiment. A poorly motivated respondents may do worse in the results3. Control * One important characteristics of a good research design is to minimize the influence or effect of extraneous variables. * The technical term 'control' is used when we design the study minimizing the effect of extraneous independent variable. * In experimental researches; the term ‘control’ is used to refer to restrain experimental conditions.Sampling Design and ProcedureSome fundamental definitions Population: Is the theoretically specified aggregation of survey elements from which sample is actually selected. Sampling Frame: Is the list of elements from which the sample is drawn Sample: A subset or some part of a larger population Sample design: is a definite plan for obtaining a sample frame Sampling: Is the process of sample to make conclusion about the whole population. Element: Is unit from which information is collected Statistic: Is a characteristic of a sample Parameter: Is a characteristic of a population.Steps in Sample Design 1. Define target population Clearly identify the specific group or universe you want to study and generalize your findings to. Once the population has been defined, a choice must be made between census and sample Census is a complete enumeration of all elements in the population Sample is a group of people who have been selected from a larger populationCensusAdvantages of census * Reliability * Detailed information Limitation of census ° Expensiveness ¢ Excessive time and energy2. Determine the sampling frame * Create a list of all the individual units within your target population from which the sample will be drawn. This is also known as the source list or sampling frame. ° Example: list of registered voters, List of students from rosters Sampling Frame = iin bodepe Dopepe bopepe bodepe bo Do peCont'd 3. Select the sampling technique * Select an appropriate sampling technique ie. random sampling vs non random sampling depending on your research goals, resources and time, and desired accuracy. 4, Determine the sample size * Calculate the appropriate sample size based on your goals, required precision, confidence level, and available resources.Cont'd 5. Select data collection method * Choose the method for collecting data from the sample, such as surveys i.e. questionaries, interviews, or observations. 6. Analyze and interpret data * After collecting data, use appropriate statistical methods to analyze it. * Draw conclusions from the analysis that can be generalized to the larger target population.Sampling Techniques Sampling techniques are basically of two types namely, non-probability sampling and probability sampling. Probability sampling give equal chance that each element of the population will be included in the sample. Non-probability sampling technique does not give equal chance that each element of the population will be included in the sample. Both have its advantage and disadvantagesProbability Sampling Probability sampling uses random selection, giving every member a known chance of being chosen, which allows for strong statistical inferences about the population Chance selection eliminates the bias because this process is random. Probability sampling is the most preferred type of sampling because of the following characteristics * sample units are not selected based on the interest of the researcher + Each unit of the population has some known probability of entering the sample Simple random sampling, systematic sampling, stratified sampling, and cluster sampling are examples of probability sampling1. Simple random sampling * Simple random sampling is a method of selecting a sample from a population where every individual has an equal and independent chance of being chosen. * It is equivalent to lottery system of samplingSteps for simple random sampling Define the population: Clearly determine the group you want to study. Determine the sample size: Decide how many individuals you need in your sample to get accurate results. Create a sampling frame: Make a list of every individual in the population. Assign random numbers: Assign a unique number to each member of the population. Randomly select the sample: Use a random number generator to select numbers until you reach your desired sample size.* For example: a company with 1,000 employees assigning each a number and using a random number generator to select 100 of them for a survey. Each employee has an equal chance of being chosen, making it a fair representation of the entire workforce feroisnano) Simple Random Sampling2. Systematic sampling This method involves selecting every n'‘> member of the population after a random starting point is chosen A probability sampling method where researchers select sample units from a larger population at a predetermined, regular intervalSample sndeedeadsSteps to perform systematic sampling Define the population: Identify the entire group from which you want to select a sample. Determine the sample size: Decide how many members you want in your sample. Calculate the sampling interval (k): Divide the total population number by your desired sample size. Formula: k=Population number/Sample Size Choose a random starting point: Select a random number between 1 and k to be your first selection. Select the sample: Start with the randomly chosen number and select every k-th member from the list until your desired sample size is reachedExample * Lets say you are a small business owner with 2,000 customers and you wanted to select 200 of them for a questioner using systematic sampling using the above steps. * Step 1. Define your population —Let’s say you have a list (or a picture) of 2000 people, numbered from 1 to 2000. * Step 2. Determine the sample size: Decide how many members you want in your sample — Lets say 200Cont'd * Step 3: Calculate the sampling interval k — Formula: k=Population Size/Sample Size —K = 2000/200 = 10 . So, you will select every 10th person. Step 4. Choose a random starting point — Pick a random number between 1 and 10. — Let’s say you randomly choose 6. Step 5: Select the sample: * you start at person #6, then pick every 10th person after that: 6, 16, 26, 36, 46, .....,19966,3. Stratified sampling a probability sampling technique where a population is first divided into homogeneous (same characteristics) subgroups, or strata based on certain characteristics that are relevant to the study —These characteristics could be age, gender, income level, geographic location, or any other factor deemed significant. Samples are then randomly selected from each subgroup to ensure proportional representation.Stratified Sampling ; i a » i dvit tfWe use Stratified sampling when the population is heterogeneous (not homogeneous) * homogeneous Population is where all items in the sample share similar or identical traits that are relevant to the research, such as age, gender, income level, educational status , location , size etc... * heterogeneous Population is where all items in the sample have diverse characteristics that are relevant to the research, such as age, gender, income level, educational status , location , size etc...IF THE POPULATION IS HOMOGENEOUSIF THE POPULATION IS HETEROGENEOUS Mi AeSteps in Stratified sampling * Step 1: Define your population and subgroups. — Identify the specific subgroups you want to ensure are included in your sample, such as by age, gender, income, or education level. * Step 2: Divide the population into strata (Stratification) — Separate the entire population into these mutually exclusive and exhaustive subgroups (strata). * Step 3: Decide on the sample size for each stratum. — Determine how many individuals to sample from each subgroup. This is often done proportionally to the subgroup's size in the total population.Cont'd Step 4: Randomly sample from each stratum. * Use a simple random or systematic sampling method to select a random sample from each individual stratum Step 5: Combine the samples. * The samples from each stratum are combined to form the final, representative sample.Example 1 Imagine a company wants to survey its 1,000 employees about their job satisfaction. The company is 80% female and 20% male. Stratification: The population is divided into two strata: Female employees and Male employees. Proportional sampling: To create a sample of 100 employees that reflects the company's gender balance, the researcher would randomly select 80 women (80% of 100) and 20 men (20% of 100) from their respective groups.Example 2 * Scenario: A public health researcher wants to study the effectiveness of a new drug against diabetes in a city from a population of 100 people. The researcher wants to ask a sample of 30 people. The population might have different responses among different age groups. Hence, he has divided the population in to strata of age group.Tot Age group 0-19 20-39 40-59 >=60 al Number of diabetic people 10 20 40 30 100 10/100*30 20/100*30 40/100*30 30/100*30 Sample Size =3 =6 =12 =9 304. Cluster Sampling * A probability sampling method where a population is divided into groups (clusters), and a random sample of these clusters is selected to be studied ¢ This technique is particularly useful for large or geographically dispersed populations because it can be more cost-effectiveExample Assume You are interested in the average reading level of all the seventh-graders in your city. It would be very difficult to obtain and collect data from all schools across the city. However, you can easily obtain a list of all schools and collect data from a subset of these. You thus decide to use the cluster sampling method. Define your population In your reading program study, your population is all the seventh-graders in your city.* Divide the population in to cluster (district or kebeles) — You cluster the seventh-graders by the school they attend. To cover the whole population, you need to include every school in the city. Assume there are 8 kebeles * Randomly select cluster(kebels). — You assign a number to each school and use a random number generator to select a random sample. — You choose the number of clusters based on how large you want your sample size to be — Assume you select 4 kebels randomly* Collect data from cluster * You test the reading levels of every seventh- grader in the schools that were randomly selected for your sample. 3G Cluster Random Sampling Define the Cluster the Randomly Collect data population population select clusters from clusters thea, a @ a a uray = => ay > ary we ap an ; é A @® ai an2. Non-probability sampling Non-probability sampling uses non-random selection based on convenience or other criteria, is faster and cheaper Non-probability sampling can lead to biased samples and limits the ability to generalize findings Uses non-random selection. Is faster and less expensive. Can lead to biased samples and does not allow for strong statistical inferences about the entire population Convenience sampling, Judgmental/purposive sampling, Snowball sampling and Quota sampling are examples of non-probability sampling. . Quota sampling Quota sampling is a non-probability sampling method where a researcher sets quotas for subgroups within a population based on characteristics like age or gender, and then non-randomly selects participants to meet those quotas It is similar to stratified sampling but lacks the random selection from each subgroup For example, a study might set a quota for 50% male and 50% female participants to ensure gender representation.Purposive (judgmental) sampling * In this approach the investigator has complete freedom in choosing his sample according to his wishes and desire The researcher uses their own judgment to select participants who they believe are the most suitable for the study's purpose. This method relies heavily on the researcher's expertiseConvenience sampling * This is a "hit or miss" procedure of study. * Participants are selected based on their easy availability and accessibility, such as stopping people in a mall to ask them questions. * It is fast and inexpensive but can lead to biased results.SampleSnowball sampling * Also known as referral sampling, this method involves initial participants recruiting others from their networks to participate in the study. * It is useful for reaching hard-to-find or niche populations but can create a sample that is biased due to the social circles of the initial participants * It is also known as Multiplicity sampling or Multi-stage Sampling.S NOWBALL SAMPLINGSnowball sampling badd 1k AO