PR2 Q2 Reviewer PDF

PR2 Q2 Reviewer Quantitative Research Designs Research design is defined as the logical and coherent overall strategy that the researcher uses to integrate all the components of the research study (Barrot, 2017, p 102). In quantitative research, you are going to have a great deal o...

PR2 Q2 Reviewer Quantitative Research Designs Research design is defined as the logical and coherent overall strategy that the researcher uses to integrate all the components of the research study (Barrot, 2017, p 102). In quantitative research, you are going to have a great deal of abstraction and numerical analysis. According to Fraenkel and Wallen (2007, p 15), the research designs in quantitative research are mostly pre- established. Hence having an appropriate research design in quantitative research, the researcher will have a clearer comprehension of what he is trying to analyze and interpret. The research problem dictates the research design. Types of Quantitative Research Design 1. Descriptive Research When little is known about the research problem, then it is appropriate to use descriptive research design. It is a design that is exploratory in nature. This design is best used when the main objective of the study is just to observe and report a certain phenomenon as it is happening. 2. Correlational Research The main goal of this design is to determine if variable increases or decreases as another variable increases or decreases. This design seeks to establish an association between variables. It does not seek cause and effect relationship, and like descriptive research; it measures variables as it occurs. 3. Ex Post Facto Research If the objective of the study is to measure a cause from a pre-existing effects, then Ex Post Facto research design is more appropriate to use. In this design, the researcher has no control over the variables in the research study. Thus, one cannot conclude that the changes measured happen during the actual conduct of the study. 4. Quasi-Experimental Research The term means partly, partially, or almost. This research design aims to measure the causal relationship between variables. The effect measured is considered to have occurred during the conduct of the current study. There is a manipulation of a variable, but members of the group are not assigned randomly. 5. Experimental Research This research design is based on the scientific method called experiment with a procedure of gathering data under a controlled or manipulated environment. It is also known as true experimental design since it applies treatment and manipulation more extensively compared to quasi-experimental design. Random assignment of subjects or participants into treatment and control group is done increasing the validity of the study. Quantitative Sampling A population is the entire set, actual or conceptual, of subjects in a survey or study for which the question of interest is asked. It is the entire set of subjects about which the statistician wants to draw conclusions. The population must be clearly defined – including the time frame – for a sample to be drawn from it. Sampling errors often occur because the population of interest is not clearly defined at the onset of a study. Sampling – refers to the process of systematically selecting individuals, units, or groups to be analyzed during the conduct of study Generalizability – refers to the extent your findings can be applied in other contexts Three Ways to Determine the Sample Size for your Study: 1. Heuristic Approach - a term normally used in qualitative studies as a research approach that utilizes introspection. In quantitative research, it refers to the rule of thumb for the sample size used in a study. Lunenburg and Irby (2008 in Barrot, 2017): 2. Literature Reviews - you may want to read studies similar to yours and check the sample size that they used. These studies can serve as a reference in proving the validity of the sample size that you plan to use. Using Formula i.e. Slovin’s Formula is used to calculate the sample size (n) given the population size (N) and a margin of error (e). It's a random sampling technique formula to estimate sampling size It is computed as n = N / (1+Ne2). whereas: n = no. of samples N = total population e = error margin / margin of error 3. Probability Sampling/Random Sampling – involves the selection of a group of participants from a larger population by chance. From a population, samples are drawn. Types of Random Sampling in Quantitative Research A. Simple Random Sampling It is a way of choosing individuals in which all members of the accessible population are given an equal chance to be selected. There are various ways of obtaining samples through simple random sampling. These are fish bowl technique, roulette wheel, or use of the table of random numbers. This technique is also readily available online. Visit this link https://www.randomizer.org/ to practice. B. Stratified Random Sampling The same with simple random sampling, stratified random sampling also gives an equal chance to all members of the population to be chosen. However, the population is first divided into strata or groups before selecting the samples. The samples are chosen from these subgroups and not directly from the entire population. This procedure is best used when the variables of the study are also grouped into classes such as gender and grade level. C. Cluster Sampling This procedure is usually applied in large-scale studies, geographical spread out of the population is a challenge, and gathering information will be very time-consuming. Similar to stratified random sampling, cluster sampling also involves grouping of the population according to subgroups or clusters. It is a method where multiple clusters of people from the chosen population will be created by the researcher in order to have homogenous characteristics. For example, a researcher would like to interview all public senior high school students across Mindanao. The clusters will be selected to satisfy the plan size. In the given example, the first cluster can be by region, the second cluster can be by division, and the third cluster can be by district. Another way of doing cluster sampling is illustrated on the figure on the right side.  D. Systematic Sampling This procedure is as simple as selecting samples every nth (example every 2nd, 5th) of the chosen population until arriving at a desired total number of sample size. Therefore, the selection is based on a predetermined interval. Dividing the population size by the sample size, the interval will be obtained. For example, from a total population of 75, you have 25 samples; using systematic sampling, you will decide to select every 3rd person on the list of individuals. Instrument Validity and Reliability Research instruments are tools used to gather data for a particular research topic. Example of instruments are performance tests, questionnaires, interviews, observation checklists Characteristics of a Good Research Instrument 1. Concise: Have you tried answering a very long test, and because of its length, you just pick the answer without even reading it? A good research instrument is concise in length yet can elicit the needed data. 2. Sequential: Questions or items must be arranged well. It is recommended to arrange it from the simplest to the most complex. In this way, the instrument will be more favorable to the respondent’s answers. 3. Valid and reliable: The instrument should pass tests of validity and reliability to get more appropriate and accurate information. 4. Easily tabulated: Since you will be constructing an instrument for quantitative research, this factor should be considered. Hence, before crafting the instruments, the researcher makes sure that the variable and research questions are established. These will be an important basis for making items in the research instruments. Ways in Developing Research Instruments 1. Adopting an instrument from the already utilized instruments from previous related studies. 2. Modifying an existing instrument when the available instruments do not yield the exact data that will answer the research problem. And the third way is when the 3. Researcher made his own instrument that corresponds to the variable and scope of his current study. Validity refers to the degree to which an instrument measures what it is supposed to measure. Types of Validity of Instruments 1. Face Validity. It is also known as “logical validity.” It calls for an initiative judgment of the instruments as it “appear.” Just by looking at the instrument, the researcher decides if it is valid. 2. Content Validity. An instrument that is judged with content validity meets the objectives of the study. It is done by checking the statements or questions if this elicits the needed information. Experts in the field of interest can also provide specific elements that should be measured by the instrument. 3. Construct Validity. It refers to the validity of instruments as it corresponds to the theoretical construct of the study. It is concerning if a specific measure relates to other measures. 4. Concurrent Validity. When the instrument can predict results similar to those similar tests already validated, it has concurrent validity. 5. Predictive Validity. When the instrument is able to produce results similar to those similar tests that will be employed in the future, it has predictive validity. This is particularly useful for the aptitude test. Reliability refers to the consistency of the measures or results of the instrument. Types of Reliability of Instruments  1. Test-retest Reliability. It is achieved by giving the same test to the same group of respondents twice. The consistency of the two scores will be checked. 2. Parallel Form Reliability. It is also known as Equivalent Forms Reliability. It is established by administering two identical tests except for wordings to the same group of respondents. 3. Internal Consistency Reliability. It determines how well the items measure the same construct. It is reasonable that when a respondent gets a high score in one item, he will also get one in similar items. There are three ways to measure the internal consistency; through the split-half coefficient, Cronbach’s alpha, and Kuder- Richardson formula. 4. Inter-Rater Reliability. Two raters must provide consistent results. A 0.70 coefficient value from Kappa coefficient, the most common statistical tool used for inter-rater reliability, means that the instrument is reliable. Research Intervention Research intervention or treatment pertains to what is going to happen to the subjects of the study. It covers who will receive the intervention and to what extent it will be applied to them The effect of these interventions can be tested by comparing two groups: the experimental group, also known as the treatment group, which is exposed to the intervention and the group that was not exposed to the intervention, the control group. Steps in Describing the Research Intervention Process 1. Write the Background Information. It is an introductory paragraph that explains the relevance of the intervention to the study conducted. It also includes the context and duration of the treatment. 2. Describe the Differences and Similarities between the Experimental and Control Group. State what will happen and what will not both in the experimental and control groups. This will clearly illustrate the parameters of the research groups. 3. Describe the Procedures of the Intervention. In particular, describe how will the experimental group receive or experience the condition. It includes how will the intervention happens to achieve the desired result of the study. For example, how will the special tutorial program will take place? 4. Explain the Basis of Procedures. The reason for choosing the intervention and process should clear and concrete reasons. The researcher explains why the procedures are necessary. In addition, the theoretical and conceptual basis for choosing the procedures is presented to establish the validity of the procedures. Quantitative Data Collection Techniques Quantitative data Data are pieces of information or facts known by people in this world. Appearing measurable, numerical, and related to a metrical system, they are called quantitative data. These data result from sensory experiences whose descriptive qualities such as age, shape, speed, amount, weight, height, number, positions, and the like are measurable. These quantitative data become useful only in so far as they give answers to your research questions. (Russell 2013; Creswell 2013). Techniques in Collecting Quantitative Data 1. Observation Information or facts are gathered using your sensory organs. Expressing these sensory experiences to quantitative data, you record them with the use of numbers. Example: watching patients lining up at a medical clinic, instead of centering your eyes on the looks of the people, you focus your attention on the number, weight, and height of every patient standing up at the door of the medical clinic. Direct observation - seeing, touching, and hearing the sources of data personally Indirect observation - seeing, touching, and hearing the sources of data by means of technological and electronic gadgets like audiotapes, video records, and other recording devices used to capture earlier events, images, or sounds. 2. Survey Gathering facts or information about the subject or object of your research through the data-gathering instruments of interview and questionnaire. This is the most popular data-gathering technique in quantitative and qualitative. This is done using any or both of the two following data-gathering instruments: Questionnaire – through paper questionnaires  Interview – similar to questionnaire, but done orally. Can also be done with the help of modern electronic devices (telephones, smartphones, computers, etc.) 3. Experiment It is the method of collecting data whereby you give the subjects a sort of treatment or condition then evaluate the results to find out the manner by which the treatment affected the subjects and to discover the reasons behind the effects of such treatment on the subjects. 4. Content Analysis It is the technique that makes you search through several oral or written forms of communication to find answers to your research questions. Data can come from non-book materials like photographs, films, video tapes, paintings, drawings, and the like. Three Phases in Data Collection Quantitative Data Analysis Data analysis is the process of systematically applying statistical and/or logical techniques to describe and illustrate, condense and recap, and evaluate data. Data interpretation is the process of reviewing data through some predetermined concepts, principles or related findings which will help assign some meanings to the data. It involves taking the result of data analysis, making inferences or implications on the relations studies, and using them to arrive at valid conclusion. Presenting and Analyzing Data It is based from: Statement of the Problem (SOP)/Objectives Hypothesis Research Instruments Statistical Tools Parametric and Non-Parametric Tests Parametric test – data is normally distributed. Non-parametric test - data is not normally distributed *Parametric tests are generally more powerful than non-parametric tests. That is why, if possible, the use of parametric tests is more preferred. Normal distribution - data are symmetrically distributed with no skew. Most values cluster around a central region, with values tapering off as they go further away from the center. Quantitative Observation and Interviews Quantitative research observation refers to the systematic collection and analysis of numerical data in order to understand, describe, or explain a phenomenon or relationship. In quantitative research, observations are based on measurable and quantifiable variables, allowing researchers to use statistical methods to analyze the data and draw meaningful conclusions. Tips To Help You Design an Effective Observation Checklist 1. Operationalize Variables: Clearly define and operationalize each variable you are measuring. This involves specifying how the variable will be observed and measured in concrete, observable terms. 2. Limit the Number of Items: Keep the checklist concise by focusing on the most critical variables and behaviors. This helps maintain the observers' attention and minimizes the chances of errors. 3. Provide Response Scales: Use a standardized response scale for each item to facilitate data analysis. Common scales include Likert scales, numerical scales, or categorical scales. 4. Consider Time Intervals: If the observation involves time-sensitive events or behaviors, consider incorporating time intervals into your checklist to record observations at specific points. 5. Ensure Inter-Rater Reliability: If multiple observers are involved, establish and test inter-rater reliability. This ensures that different observers are interpreting and recording observations in a consistent manner. 6. Include a Notation System: Include a notation system for observers to record additional comments or contextual information that may be relevant to the observations. 7. Prevent Observer Bias: Train observers thoroughly to minimize biases and ensure consistent data collection. Consider using double-blind observation techniques when applicable Quantitative interviews are a systematic approach to data collection, employing a structured questionnaire with closed-ended questions to elicit measurable responses. This type of interview is characterized by the use of predetermined questions that are asked in a standardized manner to ensure consistency across respondents. Tips in Designing Quantitative Research Interview Questions 1. Develop a Structured Interview Protocol: Create a standardized set of questions with predetermined response options. This ensures consistency across interviews and facilitates statistical analysis. 2. Ensure Clarity and Simplicity: Keep questions clear, concise, and easily understandable. Avoid jargon or complex language that could lead to misinterpretation. 3. Consider the Order of Questions: Arrange questions in a logical sequence. Start with easy and non-threatening questions to build rapport before moving to more sensitive or complex topics. Order of Interview Questions First set of questions – opening questions to establish friendly relationships, like questions about the place, the time, the physical appearance of the participant, or other non-verbal things not for audio recording. Second set of questions – generative questions to encourage open-ended questions like those that ask about the respondents’ inferences, views, or opinions about the interview topic. Third set of questions – directive questions or close-ended questions to elicit specific answers like those that are answerable with yes or no, with one type of an object, or with definite period of time and the like. Fourth set of questions – ending questions that give the respondents the chance to air their satisfaction, wants, likes, dislikes, reactions, or comments about the interview. Included here are also closing statements to give the respondents some ideas or clues on your next move or activity about the results of the interview 4. Consider Time Constraints: Be mindful of the time required to complete the interview. Keep it within a reasonable duration to minimize respondent fatigue. 5. Train Interviewers: If multiple interviewers are involved, provide training to ensure consistency in administering the interview and recording responses. Survey Questionnaires and Tests A survey questionnaire is a structured set of questions designed to gather information from individuals or groups of people. The questionnaire typically consists of a series of questions that respondents answer based on their experiences, preferences, beliefs, or other relevant factors. Tips in Designing an Effective Survey Questionnaire 1. Start with Demographics: Include demographic questions at the beginning of the survey to gather information about respondents' characteristics (e.g., age, gender, education, income). This helps in segmenting and analyzing data. 2. Avoid Leading Questions: Formulate questions in a neutral and unbiased manner to avoid influencing respondents' answers. Ensure that questions are not leading them toward a particular response. 3. Ensure Mutually Exclusive and Exhaustive Options: Provide response options that cover all possible answers without overlap. This ensures that each respondent can choose a single, accurate response. 4. Use a Mix of Question Types: Incorporate a variety of question types, including multiple-choice, Likert scales, yes/no, and open-ended questions. This adds depth to your data and caters to different respondent preferences. 5. Pilot Test Your Questionnaire: Before administering the survey to your target audience, conduct a pilot test with a small group. This helps identify any issues with wording, question flow, or ambiguity. 6. Consider Question Order: Arrange questions in a logical order to create a smooth flow. Start with easy, non- threatening questions and progressively move to more complex or sensitive topics. 7. Include a Mix of Positive and Negative Statements: When using Likert scales or similar rating systems, include both positive and negative statements to balance the survey and reduce response bias. 8. Provide an "Other" or "Not Applicable" Option: Include an option for respondents to select if none of the provided choices apply to them or if a question is not applicable. 9. Keep the Survey Length Manageable: Respect respondents' time by keeping the survey concise. Long surveys may lead to fatigue and reduced data quality. 10. Tailor your questionnaire to the mode of administration (e.g., online, phone, in- Consider the Mode of Administration: person). The format can impact question wording and response options. Quantitative tests and assessments are designed to measure specific characteristics or abilities in a numerical or quantifiable way. Tests are used to measure knowledge and skills. If it measures knowledge, it is usually through a pen-and-paper test while performance-based test is used to measure skills. The common examples of quantitative tests and assessments are intelligence tests, achievement tests, aptitude tests, personality tests, behavioral assessments, numeracy tests, etc. Tips in Choosing Quantitative Tests and Assessments 1. Clarify Assessment Purpose: Clearly define the purpose of the assessment. Is it to measure knowledge, skills, application, critical thinking, or a combination of these? Understanding your objectives will guide your choice of assessment. 2. Evaluate Test Validity and Reliability: Assess the validity and reliability of the test. Does it measure what it intends to measure? Does it produce consistent results over time and across different settings? 3. Choose Appropriate Test Format: Select a test format that aligns with your goals and the type of knowledge or skills you want to assess. Common formats include multiple-choice questions, true/false statements, shortanswer questions, and numerical problems. 4. Consider Time Constraints: Evaluate the time available for the assessment. Choose a test format and length that can be completed within the allotted time without compromising the quality of responses. 5. Think About Accessibility: Consider the accessibility of the assessment for all students, including those with disabilities or diverse backgrounds. Choose formats and tools that accommodate various learning styles and abilities. 6. Check Test Security Measures: If security is a concern, especially for high-stakes assessments, choose assessments and platforms that incorporate security features to prevent cheating and maintain the integrity of the results. 7. Look for Standardization: Standardized tests follow specific procedures during administration and scoring, allowing for a fair and consistent comparison of results. Consider whether a standardized approach is necessary for your context. 8. Seek Recommendations and Reviews: Consult colleagues, educational experts, and reviews to gather insights into the effectiveness and appropriateness of different assessments. Learning from others' experiences can inform your decision-making. Results and Discussion Results are the findings of your study after doing a rigorous analysis of the data that you collected through the use of research instrument. Its presentation follows the sequence of your specific research questions expressed in topic forms, and they serve as subheading. Meanwhile, discussion reviews the salient findings of your research vis-à-vis the previous studies you outlined or presented in the Introduction section of the paper. You may discuss the similarities between your results and the literature or studies you surveyed. You may also discuss the differences between your results and the literature or studies you surveyed. Before writing this section, Rewrite Chapters 1-3 before or after data analysis and before writing Chapter 4. Rewrite Chapters in past tense, wherever applicable, and make corrections for actual data collection and data analysis procedures. In writing this chapter, Label section headings based on research questions and follow their sequence Determine parts of the data you collected that focused on each of the variables. One subsection should be devoted to presenting data relevant to each variable. Put greater emphasis on significant results. Results that are sidelights should not receive equal weight. Tables and Graphs A results section include tables, figures, and detailed explanations about the statistical results Quantitative data are organized & summarized in tables and figures. Information shown in tables and figures is elaborated in the text. Non-prose materials/non-textual forms are graphic or visual representations of sets of data or information. Information in these materials is outlined in a way that others can easily understand between the variables being examined in the study. Types: 1. Table. Non-prose materials that help condense and classify information using columns and rows. Box heads - headings on the top and Stubs – heading on the far-left columns (title, column caption, row caption, body) 2. Graphs. Unlike tables, graphs do not merely list down the collected data with respect to a certain category. Instead, they focus on immediately representing how a change in one variable relates to another. c. Circle. Shows the relationship of parts to a whole, usually in percentages and proportions. Also known as pie chart. Test of Difference Two Branches of Statistics Descriptive Statistics It is the term given to the analysis of data that helps describe, show or summarize data in a meaningful way that patterns might emerge from the data. It presents the data in a more meaningful way, which allows simpler interpretation of the data. It involves tabulating, depicting, and describing the collected data. Common Statistical Tools for Descriptive Statistics  Frequency distribution  Percentage  Mean  Mode  Standard Deviation Inferential Statistics It makes generalizations about the population through a sample drawn from it. includes hypothesis testing and sampling  It involves with a higher degree of critical judgment and advanced mathematical models such as parametric and non-parametric statistical tools. Test of Difference Independent Sample t-test Does right-handedness or left-handedness affect the performance of students in Mathematics? Random samples of students from a class are given an exam, and the results are compared. The results are as follows: Analysis of Variance Three sections of the same mathematics subject are taught by three teachers. The exam scores were recorded as follows: Since the computed F-value (41.46) is greater than the critical value (3.80), reject the null hypothesis. Thus, there is a significant difference between the groups. Test of Relationship Pearson Product-Moment Correlation Coefficient Is age and glucose level a direct correlation? Calculate the Pearson’s correlation coefficient and interpret the result. Chi-square Test Determine whether a statistically significant association exists between smoking and being a professional athlete. The observed frequencies are presented in a contingency table shown below: Solution: Step 1. Compute the expected frequencies for each subgroup. Use the formula, Step 2. Compare the observed and expected frequencies. Use the formula, Step 3. Find the sum to obtain the test statistic. Use the formula, Since the test statistic (15.56) is greater than the critical value (3.841), reject the null hypothesis. Thus, there is a significant association between smoking and being an athlete. Drawing Conclusion Drawing Logical Conclusions from Research Findings Conclusions represent inferences drawn from the findings of the study. It should give the learning impression and it should summarize the learnings from the study. The number of conclusions corresponds with the number of specific findings. It should not contain any number or measurements. Implications of the study are also discussed in this part. If there are tested hypotheses in the study, the rejection or acceptance of hypotheses are placed under Conclusions. Aside from making generalizations from your findings, the conclusions state then implications of your findings in terms of different aspects. This means you have to identify what areas of concern or issues in your field of study can be examined and addressed based on your findings. Kinds of Implications in the Conclusions 1. Practical implications – relate to the issues in real-life situations that can be addressed through the findings.  2. Theoretical implications – relate to the issues concerning support, contradiction, and supplementation of existing models and concepts in your field of study. It can also point out how your findings can pave the way for new studies in the field. 3. Methodological implications – relate to the issues concerning materials and processes in research. Strategies for Writing an Effective Conclusion 1. Conclusions are intertwined with the Introduction Introduction: Reproductive health education is one area of research that should be tackled in schools. High school students should be exposed to innovative ways of disseminating and communicating issues or reproductive health education of high school students shall be the focus of the study Conclusion: The high school student respond nets are not fully aware and have little knowledge in reproductive health education on the following areas: social and gender related issues, family planning and other clinical services. 2. Conclusions are inferences and generalizations based upon the findings. Example: based on a research study on “Factors Affecting the Career Choices of High School Students” two (2) conclusions can be drawn from the findings of the study: Conclusion 1: Males prefer technology-based courses while females prefer business-related courses. Conclusion 2: There is a significant difference between the career choices of male and female high school respondents 3. Conclusions should specifically answer the questions posted in the “Statement of the Problem” of your study. Example 1: If the profile of the respondents will be used for test for variation on other measures, here are some examples of possible conclusions: Conclusion 1: Majority of the respondents are aged 12-16 and mostly are males. Conclusion 2: Respondents of the study are mostly Catholics. Conclusion 3: Among the profile of the respondents, age and gender are significantly related to achievement in mathematics. 4. Conclusions should contain facts or actual results from the research. Conclusions should never be based on implied or indirect implications of the findings. Example: In a research study on the “Evaluation of Mainstreaming Special Education (SPED) for Visually Impaired of DepEd, results indicated that teachers are not majors of SPED and that the facilities are inadequate to support the program. Other Guidelines in Writing Conclusions 1. Conclusions should be clearly, concisely, and briefly stated. 2. Conclusions should be original and accurate. 3. Conclusions should not introduce new arguments, new ideas or information not related to your research study. 4. Conclusions should leave the reader with an interesting final impression. 5. Conclusions should refer only to the subjects or population of your research study. 6. Conclusions should contain categorical statements. Refrain from using words like perhaps or maybe or those words that would imply unresolved issues. 7. Conclusions should not contain apologetic statements for unresolved problems in the study. Things to Avoid in a Conclusion 1. Avoid beginning your conclusion with statements like "in conclusion" or "in summary," as these basic statements can come across as redundant. 2. Do not include completely new information in your conclusion. 3. Don't wait until your conclusion to state your thesis. 4. Steer clear of rambling and be concise and straightforward as possible. Stick to the implications of your research rather than the methodologies and results of your studies (which should be in the body of your paper). 5. Resist the urge to apologize if you have doubts regarding your research paper. Formulating Recommendation Recommendations typically plays a vital portion of a research study. Formulating recommendations is easy, provided the findings and conclusions have been fully explained and completed. Recommendations are based on the findings of the study. They should not be based on your own beliefs or biases. They should not be too broad to lose its relevance to the exact topic of your research study. Characteristics of Recommendations 1. Relevance to the Study - It should be related to your study. 2. Logical Reasoning - It should be well thought with valid reasons. 3. Feasibility and Attainability - It is practical, workable, and achievable. Purpose of Recommendations 1. Policy recommendations. Example: Teachers should be encouraged to conduct research work and collaborate with other teachers in the preparation of modules, guides, and other instructional materials. 2. Recommendations for future research directions. Example: More research on the effects of playing computer games must be conducted in the future. 3. Recommendations to some problems discovered in the research study. Example: Hire more qualified teachers in the area of science and technology. 4. Recommendations for improvement or change. 5. There may be recommendations for the continuance of a good practice. How to Write Recommendations Brief – write concisely; any reason for recommendation should only be given if necessary. Clear – do not be ambiguous as to how the suggestion should be implemented. Precise – vague recommendations usually result from insufficient research or analysis.