Practical Research 2 PDF

PRACTICAL RESEARCH 2 1ST SEMESTER – QUARTER 2 – MADE BY COFFEE BEAR COVERAGE SAMPLING IN QUANTITATIVE RESEARCH Sampling –...

PRACTICAL RESEARCH 2 1ST SEMESTER – QUARTER 2 – MADE BY COFFEE BEAR COVERAGE SAMPLING IN QUANTITATIVE RESEARCH Sampling – is the process of obtaining the participants of 1. Writing Research Methodology a study from a larger pool of potential participants termed Quantitative Research Design as the population. Research Locale One of the characteristics of quantitative research is using Population and Sampling representative samples. This is essential so that the Research Ethics findings or results of the study can be generalized and Instrumentation in Quantitative Research that conclusions can be consider valid. 2. Data Analysis There are two ways of generating samples in quantitative study: a. Probability Sampling (Random Sampling) b. Non-probability Sampling (Non-random DECISION TREE IN SELECTING A QUANTITATIVE Sampling) RESEARCH DESIGN PROBABILITY SAMPLING METHODS ❖ Simple Random Sampling Also known as fishbowl technique. It allows members of the population to have an equal chance of being selected as a member of the sample. Respondents are randomly selected from a larger group. Sampling procedure: 1. Assign a number to all members of a population. 2. Randomly select or draw a predetermined number by using a table of random numbers. ❖ Stratified Random Sampling It involves dividing the population into homogeneous subgroups and then taking a simple random sample in each subgroup. Respondents are split into sub-grouped then randomly selected from each group. Sampling Procedure: 1. Divide the population into different subgroups or CHOOSING AN APPROPRIATE QUANTITATIVE strata (e.g., different year level, strand, course). RESEARCH DESIGN 2. Randomly select the members of the sample for each subgroup. RESEARC TIME DIMENSION ❖ Systematic Random Sampling H Retrospectiv Cross- Longitudina When simple random sampling or strata random PURPOSE e sectional l sampling is too tedious or complicated die to the Descriptive Retrospective Cross- Longitudinal large population, then a systematic random sampling , descriptive sectional, , descriptive can be employed. study (type 1) descriptive study (type Sampling procedure: study (type 3) 1. Number the units in the population 1 to N. 2) 2. Decide on the sample size (n), you need. Predictive Retrospective Cross- Longitudinal 3. Compute for the interval size, (k). , predictive sectional, , predictive k = N/n study (type 4) predictive study (type 4. Randomly select an integer between 1 to k. study (type 6) 5. Take every kth unit of the population as a 5) member of the sample. Explanatory Retrospective Cross- Longitudinal ❖ Cluster or Area Sampling , explanatory sectional, , explanatory When the members of the population are dispersed study (type 7) explanator study (type across a wide geographical region, then cluster y study 9) sampling is the preferred method. As an example, a (type 8) sampling of all areas of Manila might be very 1|EYAH PRACTICAL RESEARCH 2 1ST SEMESTER – QUARTER 2 – MADE BY COFFEE BEAR difficult. A researcher can instead randomly select a number of districts to be members of the sample. INSTRUMENTATION IN QUANTITATIVE Sampling procedure: RESEARCH 1. Divide population into cluster using geographical Instrumentation – the process of collecting data. boundaries. Instrument – the tool or device used to collect data. 2. Randomly sample clusters. The first step in constructing an instrument is to ask the 3. Randomly select units from each sampled cluster. ff. questions on data collection: 1. How will the data be gathered? NON-PROBABILITY SAMPLING 2. When will the data be gathered? ❖ Accidental or Convenience Sampling 3. Where will the data be gathered? Participants are sampled according to what is 4. How will the data be analyzed? conveniently available. Example: RESEARCH INSTRUMENTS 1. A psychologist samples his or her own clients Research instrument – a tool or device that researchers since they are readily available. use to collect, measure, and analyze data from 2. A market researcher asks volunteers in a mall to participants in a study. be interviewed. It can take many forms depending on the type of research ❖ Quota Sampling and the data being collected. A predetermined number or percentage of the The purpose of a research instrument is to gather population is sampled. accurate, reliable, and valid data that will help answer the Example: research questions or test the hypotheses. 1. For example, you know that in a given population, there are 60%. In quota sampling, COMMON TYPES OF RESEARCH INSTRUMENTS you will select samples non randomly until you 1. Surveys/Questionnaires reach 40% women (4 out of 10). A set of written or digital questions designed to ❖ Snowball Sampling gather information from respondents on their Participants identify other potential participants to be opinions, behaviors, or characteristics. It can include: included as samples. o Open-ended questions: Allow participants to Example: answer in their own words. 1. Students belonging to a study group can o Close-ended questions: Provide specific answer recommend members of the group to be options (e.g., multiple-choice, yes/no, Likert participants on a research about intrinsic scales). motivation to study. Scale Left- Left of Center Right Right ❖ Purposive Sampling most center of -most Also known as judgmental or selective sampling. cente The researchers select participants based on their r judgment and the purpose of the study. Rating 1 2 3 4 5 Example: Satisfac Very Dissatis Neither Satisfi Very 1. If you’re studying the experiences of experienced tion dissatis fied satisfied/diss ed satisfi teachers in special education, you would fied atisfied ed purposively select teachers with long history in Quality Very Poor Fair Good Very that field. poor good Frequen Never Rarely Occasionally Freque Very cy ntly freque ntly Perform Awfull Not Work in Well Super ance y well progress bly Importa Not at Slightly Moderately Very Extre nce all importa important import mely import nt ant import ant ant Focus Much Less Maintain More Much less focus focus more focus focus 2|EYAH PRACTICAL RESEARCH 2 1ST SEMESTER – QUARTER 2 – MADE BY COFFEE BEAR 1. Descriptive Statistic – procedures that researchers use to 2. Interviews describe data. It can also include ways on how to visually A verbal method where the researcher asks present data using pie charts and bar graphs. participants a series of questions to gain deeper o It is helpful in the analysis and interpretation of insights into their experiences, opinions, or raw data by using the following methods. behaviors. Interviews can be: ▪ Frequency counts o Structured: All participants are asked the same ▪ Percentages set of questions. ▪ Measures of central tendency (e.g., mean, o Semi-structured: Allows for flexibility in the median, and mode.) questions based on responses. ▪ Measures of variability (e.g., range, standard o Unstructured: More conversational, with no set deviation, and variance). list of questions. ▪ Use of pie charts and bar graphs. 3. Observations A. Frequency Counts – frequency or counts tell you Researchers observe and record behaviors, actions, or how many time something had occurred. In the events in a natural or controlled environment. This context of survey instrument, it refers to how many method is often used in social science and of the participants belong to a certain category of a psychology research. given demographic variable, it also tell how many 4. Tests times a certain item is rated according to the scale Standardized tests or assessments used to measure used in the instrument. specific variables such as knowledge, skills, B. Percentage – will simply tell you the proportion our intelligence, or personality traits. of the total based on 100. - Percentage Formula: When selecting a research instrument, consider: 1. Validity – does the instrument measure what it is supposed to measure? 2. Reliability – is the instrument consistent in measuring data over time? - Units of parts and whole must be consistent so 3. Suitability – does it align with your research that they will cancel each other out. questions and objectives? C. Frequency Percentages Distribution 4. Practicality – is it feasible to use in terms of time, cost, and participant availability? RESEARCH ETHICS Refers to a set of guidelines that ensure research is conducted in a responsible and morally sound manner, respecting the rights, dignity, and welfare of all participants. Informed consent - Participants must be fully informed about the research's purpose, procedures, risks, and benefits, and they must voluntarily agree to participate. They should also know they can withdraw at any time. Confidentiality and Privacy - Researchers must protect participants' personal information and ensure that their identities remain confidential unless explicitly agreed upon. Integrity - Researchers must conduct their work honestly, report data accurately, and avoid fabricating, falsifying, or misrepresenting results. DATA ANALYSIS Quantitative Data Analysis – a process of analyzing and interpreting numerical data. It is about analyzing data that is number based or data can be easily converted into numbers without losing any meaning. Analyzing Quantitative Data: 3|EYAH PRACTICAL RESEARCH 2 1ST SEMESTER – QUARTER 2 – MADE BY COFFEE BEAR D. Measures of Central Tendency – it describes how data 3. Variance – simply the square of the standard sets are grouped together around a central value. It gives deviation. a picture of what is typical in the sample or what are the - Formula: most apparent and representative characteristics of the sample. Most common measures of this are the mean, median, and mode. 1. Mean or average, is the most common form of F. Pie Charts & Bar Graphs reporting central tendency. Obtaining an average is 1. Pie Chart – circular graph consisting of slices or simply taking the sum of all the answers divided by wedges that represents a percentage of the whole. the total. 2. Bar graphs – uses rectangular bars to represent data. - Formula: Each bar’s length is proportional to the value it represents, and bars can be plotted vertically or horizontally. PART II: Content 2. Inferential Statistic – the mathematics and logic of how characteristics of the sample can be generalized to a larger population. This made possible by testing the hypothesis. 2. Median – is the midpoint of a distribution. This is - used to determine weather you can infer the characteristics useful if you are reporting scores on an achievement of the population from the sample’s characteristics. test. To determine the median, arrange the scores from lowest to highest then select the middle value to ❖ Hypothesis Testing – the logical arguments and represent your median value. procedures associated with the informed decisions in research. - Steps in statistical hypothesis testing: 1. State the null hypothesis. Null hypothesis (Ho) is a statement of nullity. This means that at the onset of the study, the researcher is stating that there is no difference, correlation, or association between the variables being investigated. 2. State the alternate hypothesis. Defined as 3. Mode – simply the most reported case or incident, or the hypothesis that a researcher is willing to the most repeated value in the data set. support if the null hypothesis is rejected. E. Measure of Variability – it deals with the spread of a measurement. This is equivalent to asking how far apart are the scores or measurements from each other or from the mean. 1. Range – simply the difference between the highest value to the lowest value in a given set of measurement. It simply tells the maximum and minimum value for a set of measurement. - Formula: Range = Maximum Value – Minimum Value 2. Standard Deviation – tells you how far the 3. Set the statistical significance (alpha level) measurements are from the mean or how a given set of scores deviates from the mean. - Formula: 4|EYAH PRACTICAL RESEARCH 2 1ST SEMESTER – QUARTER 2 – MADE BY COFFEE BEAR To determine T-test of Mann-Whitney - the practical statistical procedures used in whether the independent U test hypothesis testing that assist researchers in means of samples reaching a decision are called test of independent significance. These are equivalent in determining samples are the test statistic to use. significantly different 4. Collect data. To determine T-test for Wilcoxon signed 5. Calculate the test statistic. After setting up whether the dependent rank test the significance level, the next step is to means of samples calculate a test statistic to the data collected. dependent This test can be: sample (paired) a. Z-test – comparing the mean of the are significantly sample from the mean of the population. different b. T-test – comparing the means of To test for the Pearson Chi-square test independent and dependent samples. association correlation, c. Testing for the significance of r (i.e., a between two or spearman rank correlation coefficient) more variables order correlation coefficient, and 6. Draw conclusion about null hypothesis. point biserial The last step in hypothesis testing is to either correlation reject or support the null hypothesis. coefficient ▪ In inferential statistics, this is done by comparing the test statistic to the critical Descriptive vs. Inferential Statistics values found in standard statistical tables. Aspects Descriptive Inferential ▪ If the test statistic is greater than the Statistics Statistics critical value, then we reject the null hypothesis. Purpose Summarize and Make inferences Inferential Statistic is a more appropriate and meaningful describe data about a data analysis to use for testing the null hypothesis. population Not limited to experimental or quasi-experimental Key methods Central Hypothesis research design. tendency, testing, Can be used to test: variability, significance tests o Relationship between two variables (test of charts significance of the correlation coefficient). Focus Data at hand Generalizations o Degree of association between variables (chi- beyond data square test). o Comparing the characteristics of sample from Data types Any data Requires population (significance test for a single mean) structure sampling and Parametric vs. Nonparametric Test often CHARACTERISTICS OF PARAMETRIC AND assumptions NONPARAMETRIC TESTS examples Mean, median, t-tests, chi- characteristics Parametric test Nonparametric mode, standard square tests, test deviation correlation Distribution of Normally Not normally sample distributed distributed Sample size Large Relatively small Types of data Continuous, Ordinal and interval, or ratio ranked EXAMPLES OF PARAMETRIC AND NONPARAMETRIC TESTS purpose parametric nonparametric 5|EYAH