Proposal Writing, Stages & Strategies with Examples PDF

Proposal writing, Stages and Strategies with examples A paper presented for PGD Ultrasound Course for Radiographers (Research Methods II i.e. Statistical Methods) held at Posh Hospital, No 100 Chime Avenue, New Haven, Enugu, Nigeria on 23rd July, 2019. By Charles Ugwoke Eze, B.Sc.,(...

Proposal writing, Stages and Strategies with examples A paper presented for PGD Ultrasound Course for Radiographers (Research Methods II i.e. Statistical Methods) held at Posh Hospital, No 100 Chime Avenue, New Haven, Enugu, Nigeria on 23rd July, 2019. By Charles Ugwoke Eze, B.Sc.,(Hons) {Rad.}, M.Sc., Ph.D., (Nig) (Medical Imaging). Department of Medical Radiography & Radiological Sciences, Faculty of Health Sciences & Technology, College of Medicine, University of Nigeria, Enugu Campus. Why do we embark on research proposal?  To make an original contribution to advance the existing stock of knowledge;  As a basis for sound health policy formulation;  For social and economic growth progress. Social and economic growth progress depends on quality and utilization of research evidence by government, policy makers and social groups;  To satisfy one’s intellectual yearnings for better understanding of one’s environment. That is to answer critical questions bothering the scientist or researcher;  For undertaking undergraduate and postgraduate projects;  To generate data for publications; As a hobby. Principles in Writing a Research Proposal  Address a research problem that is particularly significant in the context of Nigeria or area concerned  Internally consistent (synergy between objectives and methodology)  Includes procedures for addressing every objective  Presented in such a manner that flows logically from section to section and subsection to subsection  Easily readable with clear and concise language Criteria for selection of a Topic:  Relevance ◦ Should be a priority problem. ◦ Think of serious health problems that affect a great number of people.  Avoidance of duplication ◦ Find out whether the topic has been investigated before, either within the proposed study area or in another area with similar conditions. ◦ If the topic has been researched, the results should be reviewed to explore whether major questions that deserve further investigation remain unanswered. If not, another topic should be chosen.  Feasibility of study ◦ Consider available resources Characteristics of a Title  Should contain the name of the area where the study is to be conducted  Should be concise and not vague (must be sharp)  Not too long  Keep it short and simple (KISS)  Informative  Respond directly to a call  Should be active  Should be focused Outline of a Proposal  1.Title/Topic  2. Introduction  Background information  Problem statement  Justification/rationale/relevance to Radiography  Research questions  General & Specific Objectives  Aim (General Objective)  Specific Objectives ◦ Hypotheses  3. Literature Review  - Conceptual review  -Empirical review Outline  of a 4. Research Methods Proposal Continued. ◦ Study area ◦ Study design ◦ Study population  Inclusion criteria  Exclusion criteria ◦ Sample size determination ◦ Sampling technique ◦ Study instruments ◦ Data Collection Methods ◦ Plan for Data Management  Measurement of Variables  Statistical analyses ◦ Ethical considerations ◦ Limitations of the study  5. References  Vancouver, Harvard etc  6. Appendix ◦ Study instruments ◦ Ethical approval ◦ Work plan ◦ Others (maps e.t.c) 2. Introduction  This section goes from general (what is reported in the literature, what is done in practice) to more specific (your study).  Should contain the following sections: ◦ problem statement ◦ Rationale/justification/significance of the study ◦ objectives of the study  Statement of Problem  A concise description of the nature of the problem (the discrepancy between what is and what should be) and of the size, distribution and severity of the problem (who is affected, where, since when, and what are the consequences).  Use a literature review (original research papers and review articles) to provide this context.  Go from general to specific e.g. If you are doing a study on adherence to HIV treatment in Nigeria, start with some information on the HIV epidemic, then the treatment options and lastly the issues with adherence. And within these sections start with global evidence, then Africa, then Nigeria. Introduction continued.  Justification/Rationale for the Study  Relates to the origin/source of the topic and the importance of the problem.  A brief description of any solutions to the problem that have been tried in the past should be given, how well they have worked, and why further research is needed.  A description of the type of information expected to result from the project and a clarification of how this information will be used to help solve the problem (contribution to existing knowledge).  Again, use literature to support this.  Objectives of the Study  General objective (general aim or purpose of the study which is derived from the research topic) : ◦ What the study hopes to achieve  Specific objectives which are based on your general objective. ◦ statements of the research question(s). ◦ Should be SMART ◦ Should be:  simple (not complex),  specific (not vague),  stated in advance (not after the research is done).  Use action verbs that are specific enough to be evaluated, for example: to determine, to compare, to verify, to calculate, to describe, to establish etc.  Don’t put too many specific objectives or over-ambitious objectives that cannot be adequately achieved by the implementation of the proposal. ◦ 3 to 5 specific objectives may be okay  For each specific objective, think of the indicators to be measured ◦ Indicators are what will be measured to show whether or not objectives have been achieved Hypothesis  Testing An investigator is often required to make decisions or judgements concerning differences of various kinds. For instance, one may be assessing the effect of a  specific drug on blood pressure or the effect of consumption of a food item on blood cholesterol levels, or the effectiveness of a potential haematinic in restoring haemoglobin concentration of anaemic patients.  Usually, these judgements are made on the basis of samples drawn from parent populations which are too large to be measured directly.  In statistics, the interest is usually in two possible lines of action – to accept or reject an assumption or hypothesis.  A statistical hypothesis is thus a statement which may or may not be true concerning one or more population.  The test of a hypothesis, therefore, is a procedure for deciding whether to accept or reject the hypothesis with data relevant for doing so.  Forms of hypothesis:  There are two forms of hypothesis:  Null hypothesis (Ho), and  Alternative hypothesis (HA).  The null hypothesis (Ho) is a declaration or an assumption that there is no significant difference between two or more comparative groups or variables, i.e. whatever differences observed are purely insignificant.  Examples of null hypotheses:  -The drug being tested has no significant effect on blood pressure.  -The food consumed didn’t significantly affect blood cholesterol levels.  -The haematinic being tested has no effect on haemoglobin concentration of  the anaemic patients.  The null hypothesis is the one commonly used for most scientific investigations. Forms  of hypothesis contd. An alternative hypothesis is a declaration or assumption that there is a significant difference between two or more comparative groups or variables.  Examples of alternative hypothesis:  The drug being tested affects blood pressure.  The food consumed alters blood cholesterol.  The haematinic affects haemoglobin concentration of anaemic patients.  The alternative hypothesis does not indicate the direction of the difference; it only states that there is a difference!  The alternative hypothesis is not commonly used!  Efforts are not usually made to prove a null hypothesis.  Rather we can prove that a justification exists for rejecting an alternative hypothesis, and if we are able to do this, then we fail to reject a null hypothesis. What really constitutes a significant difference?  In its simplest form, a significant difference has been defined as ‘a difference that is large enough to make a difference’.  The judgement as to whether a difference is large enough to make a difference is determined in part by the investigator and in part by the results of the statistical test used.  It is essentially arbitrary and mechanically determined.  Statistical tests are therefore a tool which when used judiciously by the investigator will help him/her to make valid judgements.  Statistical significance must be differentiated from practical significance.  In contrast to statistical significance, practical significance is determined by the technical knowledge and insight of the investigator as he/she interprets the statistical results in terms of the pertinent variables of the experimental setting.  Consideration of practical significance enables the investigator to put into consideration the various possibilities inherent in a specific situation being investigated. Steps in Hypothesis Testing  i. State the hypothesis.  ii. Decide on the margin of error to be allowed, i.e. specify the significance level.  iii. Indicate whether a one-tailed or two-tailed test is intended.  iv. Specify the test statistic and its sampling distribution.  v. Calculate or compute the value of the test statistic.  vi. Extract the critical value.  vii. Compare the critical value with the computed/calculated value of the test statistic.  Level of Significance  The level of significance is a pre-set probability of rejecting a true null hypothesis. Put in a more technical way, it is the probability of committing a Type I error (I shall later explain what Type I error is).  -The level of significance is usually denoted by alpha (α).  -The maximum probabilities or levels of significance that has been established as acceptable rejection points are 0.05 (5%) and 0.01 (1%).  -It is therefore conventional to fix the error margin at either 5% or 1%.  -This implies that the investigator fixes the probability of making a wrong decision (i.e. rejecting a null hypothesis when it is in fact true) at α = 0.05 or 0.01.  At α = 0.05, it means there are about 5 chances in 100 that we would reject the hypothesis when it should be accepted, i.e. we are about 95% confident that we have made the right decision.  The lower the level of significance one sets for oneself, the more confident are we that we are making the right decision. What are Type I (α) and Type II (β) errors?  Type I and Type II errors are possible wrong decisions that may be made based on errors arising from computational and sampling techniques.  Type I error is committed when one is misled to reject a true null hypothesis,  while Type II error occurs when one is misled not to reject a false null hypothesis.  It should be noted that when we reduce the risk of making a Type I error, we automatically increase the risk of making Type II error, and vice versa.  -An investigator is usually more interested in the probability of a Type I error.  Good investigative procedures would involve a reasonably low level of significance in order to minimize the probability of a Type I error, and a sufficiently large sample to minimise the probability of Type II error. Critical Value of a Test Statistic  -The critical value of a test statistic is that value which is required for significance at a given level of significance.  It usually depends on the chosen level of significance and on whether the test is non- directional (two-tailed) or directional (one tailed).  -The determinant of whether a test will be non- directional or directional is the nature of the alternative hypothesis.  A non-directional alternative hypothesis states that there is a difference between the two parameters,  while a directional alternative hypothesis states that one parameter is greater than or less than the other. One-tailed and two-tailed tests!  -If the alternative hypothesis is non-directional (i.e. not specifying a direction in the difference between the parameters), a two-tailed test will be required.  -On the other hand, if the alternative hypothesis is directional (i.e. specifying a direction in the difference between the two parameters), then a one-tailed test is involved.  -Whenever we state a hypothesis, we should be able to know whether it is one or two-tailed test that is required;  Once the hypothesis gives a direction of difference, it is one-tailed, but if it does not, then it is two-tailed. Degree of Freedom (df)  -In its simplest form, the degree of freedom refers to the number of ways in which any set of scores is free to vary.  -In practice, the degree of freedom is given as: df = n – 1 (where n = size of the population or sample).  -It should be noted that the degree of freedom is tied to the size of the population or sample. Commonly used Test Statistics Z – test  The Z – test is usually adopted in testing hypothesis about the difference between two population means when the sample size is large.  Generally a sample is considered to be large if its size is equal to or greater than 30.  Otherwise, the sample is regarded as small.  The Z – test can only be used for samples or populations with known variance.  Thus in order to run the Z-test in Microsoft Excel, first run the descriptive statistics to get the variance, then use the variance obtained to do the Z –test. Instances…  Student’s t – test  This statistic is more like the Z-test, but it has the advantage that it can be used for small sample sizes.  The t – test has the major advantage of reducing Type I errors that might otherwise result from use of small sample sizes.  There are three types of student’s t – test: (i) Two-sample t – test assuming unequal variances. (ii). Two sample t – test assuming equal variances. (iii) Paired sample t – test.  The two-sample t – test assuming unequal variances is used for two data sets derived from distributions with unequal variances. Student’s t – test contd.  It is referred to as heteroscedastic t – test.  This is the commonly used t – test when there are distinct subjects in the two samples. Instances …  The two-sample t – test assuming equal variances is used for two data sets derived from distributions with the same variance.  It is referred to as homoscedastic t –test.  Since one is not always sure that the variances are the same, it is always advisable to run the test that assumes unequal variance.  Experience has shown that there is little or no difference between these two for most samples. Instances …  The paired sample t – test is used when there is a single set of subjects and the two sets of samples represent measurements from each subject before and after a treatment.  This type of t – test is commonly known as before and after t – test. Instances … Analysis of Variance (ANOVA)  Analysis of variance (ANOVA), also known as F – ratio test is used to determine the nature and scope of variances existing within and between three or more comparative samples.  It assumes that variance within groups is due to natural variations, while variance between groups is due to the treatment given.  When ANOVA is used as a test statistic, variability within the groups is as great as or greater than variability between the groups,  we would conclude that the treatment effects are non- significant.  On the other hand, if variability between groups is significantly greater than variability within groups, it shall be concluded that the effects of the treatment were significant.  There are three forms of ANOVA: (i) One way ANOVA (also known as single factor ANOVA), (ii) Two-way ANOVA (also known as two-factor ANOVA), and (iii) Repeat measure ANOVA.  one way ANOVA  The one way ANOVA is used for when there are more than two treatment groups composed of distinct subjects and when what is being analysed for is only in one dimension.  This is the most commonly used ANOVA because of its ease of interpretation.  For instance an investigator wanted to study the effects of storing blood samples under varying temperatures [refrigerator, room temperature and incubator temperature (37oC)] for 48 hours on their haematological values  – he/she had ten blood samples assigned to each of the groups.  two way ANOVA  The two way ANOVA is used for when there are more than two treatment groups composed of distinct subjects and when what is being analysed for is along two different dimensions.  Under the two way ANOVA we have what is commonly referred to as factorial design which permits the separation and evaluation of the effects of each of the two or more factors operating in a single experiment.  For instance if the investigator wanted to study the effects of varied storage temperatures and period of storage of haematological values of blood samples kept at refrigerator, room temperature and incubator temperature (37oC) for a period of three days;  he/she also has ten blood samples assigned to each of the groups.  Because of the complexities that may arise during the interpretation, it may be better to advise students to run a one- way ANOVA for storage temperatures and another one way ANOVA for period of storage – this will give exactly the same result as a two way ANOVA.  repeat measure ANOVA  The repeat measure ANOVA is an equivalent of the paired sample t – test for more than two points of measurement for one set of subjects.  Using the above instance, if period of storage is the only factor being analysed for on the samples stored at room temperature.  Haematological analysis is done on these samples at the point of collection, and at 12 hour intervals for a period of three days. Post-hoc analysis for variant means  The post-hoc analysis of variant means is necessary because an ANOVA only tells us that there is or there are no significant variations between the means of the three or more groups.  When there is a significant variation, the variant means are not usually defined by ANOVA.  Further post-hoc tests therefore help us define the specific means that are significantly different from others.  There are a variety of post-hoc tests and each is chosen for one reason or the other.  The least significant difference (LSD) method is one of the most commonly used post-hoc procedures because it is very sensitive.  Others generally used include Duncans multiple range test, Scheffe and Tukey.  There are also others such as Bonferroni, Sidak, Gabriel, Dunnet etc. Parametric statistics  The above tests of hypothesis about the difference between population means (Z –test, Student’s t – test and ANOVA) are usually referred to as parametric statistics.  They are known as parametric statistics because they are comparisons involving setting up of a level of significance, degrees of freedom and determining whether the obtained statistical value is higher or lower than the given critical value, and accepting or rejecting a given hypothesis.  In other words, parametric statistical techniques involve the estimation of parameters and testing of hypothesis concerning them.  This is the normal probability model. Non-parametric statistics  In contrast when the underlying distribution is not known, non-parametric statistics are used.  Non-parametric statistics are distribution-free tests of significance because they make no assumption about the shape of the distribution tested.  Examples of non-parametric statistics include Chi square, the sign test, Kolmogorov- Smirnon test, Wilcoxon test and the Mann- Whitney U – test.  Chi square (X2) test  The Chi square test is one which allows us to determine when or not a significant difference exists between observed data versus expected data or number of cases in each category of a variable versus total number of cases.  It is also used in testing for independence/association of two variables (testing whether one variable is dependent or not/associated or not with the other).  Chi square commonly involves an ‘A’ or ‘non A’ type of data, i.e. positive or not positive.  Example, people treated with a new drug for malaria may recover or not recover, association between skin colour and development of skin cancer, association between sex and occurrence of a specific disease (tuberculosis).  Chi square always involves a contingency table. Chi square contd.  The Chi square is a two-tailed test. It only indicates a difference or association but not the direction of the difference or association.  It should be noted that only frequency data or nominal data may be analysed using Chi square.  It is not proper to use percentages/proportions for Chi square analysis, rather such proportions/percentages should be converted to frequencies before Chi-square is applied.  Chi square can only be used when each cell in a Chi square contingency table contains a minimum expected frequency of 5.  As a rule, if the grand total in a contingency table is less than 20 or any of the observed value is less than 5, chi- square should not be used; rather Fisher’s exact test should be used. Literature Review  A SUCCESSFUL PROPOSAL STARTS WITH A DETAILED LITERATURE REVIEW  What is a literature review (LR)?  LR is an account of what has been published on your topic of interest.  Generally, the purpose of a review is to critically analyze a segment of a published body of knowledge through summary, classification, and comparison of prior research studies, reviews of literature, and theoretical articles.  In writing the LR, your purpose is to convey to your reader what knowledge and ideas have been established on a topic, and what their strengths and weaknesses are.  It provides the framework of your research investigation and summarizes the literature you studied in order to develop your research study. Why do a Literature Review?  Why do a Literature Review?  Find gaps in the literature  Avoid reinventing the wheel (at the very least this will save time and it can stop you from making the same mistakes as others)  Build on the platform of existing knowledge and ideas  Learn about other people’s working in the same field  Identify important works about your topic  Provide the intellectual context for your own work, enabling you to position your project relative to other works  Learn about opposing views  Discover information and ideas that may be relevant to your project  Identify research methods that could be relevant to your project How to do a Literature Review  Selecting Sources for the Literature Review  A good literature review requires a good literature search. ◦ an exhaustive search of the resources and information about your research topic.  Initial reading to become familiar with topic ◦ Material in reader or a review article ◦ Identify key words (and key issues)  Snowball - look for references in articles  Online search of electronic databases: ◦ Keep track of what you have searched – where & with what keywords (methods; avoid duplication) Possible Sources of Information for LR  Individuals, groups, and organizations  Published information (books, articles, indexes and abstract journals)  Unpublished information (grey literature, other research proposals in related fields, reports, records and computer data bases)  Clinic and hospital-based data from routine statistics registers  Statistics issued at National, state and LGA levels Reviewing and Writing down the Literature  Once you have identified the key resources for your topic, you then need to describe, critique and relate each source to the subject of the inquiry. ◦ This prepares you to organize your literature review logically  Work out a structure as soon as possible and take notes using that structure: ◦ Saves time ◦ Helps to develop a clear line of argument ◦ Decreases the likelihood of plagiarism ◦ Increases the potential for analysis (unpack into constituent parts; isolate main variables; determine relationship between them)  Review abstracts and/or skim whole article: ◦ Check for relevance. ◦ Identify key issues  More detailed reading and note-taking: ◦ Key issues ◦ Good quotes ◦ Source of information  Reference as you go along Structure of a LR: Create a Table of Information  Reference (Author, title, article)  Issues  Thematic area (introduction, Objectives, methods)  Remarks (special quotes) Some Helpful Reminders for Doing a LR  The process of doing a LR can be divided into two parts: ◦ the search process ◦ the writing of the review.  Establish criteria for selecting materials that will be included in the review.  Continue the search until you start noticing that you find the same sources no matter where you search. This is sometimes referred to as the saturation point in the search process.  Locate a reasonable number of sources and begin reading those sources  Arrange the materials reviewed into categories that are suggested by the material itself. (E.g. By objectives, topic, by date range, by region, etc.)  Structure the write-up of the review into three parts: introductory material, the body of the review, and a concluding section.  Check List  Must be clear that you have:  Identified the key literature relevant to the topic  Reviewed the literature thoroughly and with an ‘open-mind’  Identified the key ideas, concepts (and methods)  Recorded your sources accurately and consistently  Analysed the information in a systematic, comprehensive and relevant way Electronic Searches  Electronic databases: ◦ http://www.lib.uct.ac.za/  (Electronic Resources, Databases by platform) ◦ Science Direct ◦ EBSCO HOST (Medline, CINAHL etc.) ◦ Pub Med ◦ Popline  Websites: ◦ http://www.eldis.org/healthsystems/ ◦ http://www.phrplus.org/ ◦ http://www.worldbank.org/html/extdr/hnp/hnp.htm ◦ http://www.who.int/en/  HINARI  User Name: NIE 148  Password: 76711  User Name: NIE008  Password: 344M84  User Name: NIE038  Password: 25909  Research Methods  Most important part of the proposal.  Should include information on ◦ where the study will take place, ◦ The design of the study  information on the type of study,  the research population ◦ How to select those to be studied, ◦ Statistical methods used to calculate the sample size ◦ measurements to be taken, ◦ observations to be made, ◦ how to analyze your data.  Study Area  Description of the place where the study will be conducted: at home or in the hospital or in the health centre or in institutions  Describe in details: Location, demographic characteristics, climatic conditions, disease prevalence, social amenities etc.  In case an intervention study is conducted, you need to add a section in which you describe the interventions that the treatment and control groups received.  Study Design  Study design is often explained in one sentence e.g.  A randomized controlled trial to be conducted at the outpatient clinic of UNTH Enugu  A cross sectional descriptive study to be conducted at Nsukka LGA, Enugu State  A longitudinal study to be conducted at LUTH, Idi-Araba, Lagos State  A rough guide to research Designs No Were comparisons made? Descriptive study Yes Were subjects chosen based on the Yes Case-Control study outcome? No No Were subjects followed overtime? Cross-sectional study Yes Did the investigator make interventions? No Cohort study Yes Were the interventions assigned Yes Randomized trial randomly? No Were subjects compared with Yes Before-after study themselves? No Were intervention subjects compared Yes Randomized control with other subjects? trial No Sample Size Determination  The number of subjects planned to be enrolled. (Show how you arrived at the number you intend to use)   Sample size for one group  N = Zα2 P (1-P)  D2  Zα = significant level usually set at 95% confidence level, Zα is 1.96 (two sided)  P = Prevalence of the attribute under study  Sources of P are: literature review, Pilot study, best guess of 50%  D = Margin of error tolerated (usually set at 0.05) The minimum sample size for each group (intervention and control)  The minimum sample size for each group (intervention and control) is calculated using the sample size formula for comparison of groups as follows:  N = (Zα + Z2β)2 {n1(1-n1) + n2(1-n2)}  (n2-n1)2  Where N = minimum sample size  Zα = significant level usually set at 95% confidence level, Zα is 1.96 (two sided)  Z2β = power ie the probability of not rejecting the null hypothesis set at 80%. At power of 80%, the percentage point of the standard normal deviation (one sided) will be 0.84  n1 = Control group response i.e 32.8% (e.g. percentage of mothers who know and practice exclusive breastfeeding)  n2 = anticipated change in the study group ie 30%   N = (1.96+0.84)2 {0.328(1-0.328) + 0.30(1-0.30)}  (0.30-0.328)2  N = 120  Because of anticipated non-response of 10%, the number is increased to 132 and rounded off to 140 Sampling, sampling frame and sampling method.  Which is the larger group (sampling frame) from which the participants (sample) will be recruited for research?  How are you going to select (sampling method) the sample?  Non probability: ◦ Quota sampling ◦ Purposive ◦ Incidental/Convenience ◦ Snowballing  Probability: ◦ Simple Random sampling ◦ Systematic sampling ◦ Stratified sampling ◦ Cluster sampling ◦ Multistage sampling Common Sampling Techniques  The commonest sampling techniques adopted in research are:  (i). Simple random sampling,  (ii) Stratified random sampling,  (iii) Cluster sampling,  (iv). Systematic sampling, and  (v). Purposive sampling. Simple Random Sampling  This is the easiest and simplest of sampling techniques.  Each element has equal and independent chance of being selected.  Simple random sampling can be done by:  (i) Tossing of a coin or dice,  (ii) Use of slips of paper on which name or an identification mark of each member of the population has been written,  (iii) Use of table of random numbers,  and (iv) Use of the computer. Stratified Random Sampling  In stratified random sampling, the population is first stratified in terms of one or more variables of interest to the investigator.  Elements are then drawn randomly from each stratum of the population.  Stratified random sampling can be proportionate or disproportionate. proportionate stratified random sampling  In proportionate stratified random sampling, the elements are drawn from each stratum in such a way that the relative proportions of the strata in the resultant sample are the same as exist in the parent population.  The relative contribution of each stratum in the population is exactly its relative contribution in the sample.  Thus, a proportionate random sample shall possess specified characters in exactly the same proportion as these characters exist in the parent population. disproportionate stratified random sampling  In disproportionate stratified random sampling,  the relative proportions of the strata in the sample do not correspond to their relative proportions in the population,  such that some strata may be over- represented  or under-represented in the sample. Cluster sampling  In cluster sampling, the population or geographical area is divided into units or sections with distinct boundaries,  and then using simple random sampling, a specified number of these units or sections are drawn.  All the elements in the units or section drawn now constitute the sample.  Cluster sampling technique does not guarantee equal number of elements in each sampling unit  – a condition that can increase the bias of the resultant sample. Systematic sampling  In systematic sampling, the elements are drawn at specified intervals from a list containing all the elements in the population.  Usually the first element is randomly drawn from the first ‘n’ elements and thereafter every ‘nth’ element on the list is drawn for inclusion in the sample.  The elements can be arranged in any order – alphabetically, serially, etc.  In systematic sampling, there may be undue over- representation, higher sampling error and some bias Purposive sampling  In purposive sampling, specific elements which satisfy some pre-determined criteria are selected.  Although the specified criteria to be used are usually a matter of the researcher’s judgement,  he/she exercises this judgement in relation to what he considers will constitute a representative sample with respect to the purpose of the study.  The investigator must specify the reason for purposive sampling.  Data Collection  Describe the instrument(s) to be used for data collection (e.g. interview guide, questionnaire, checklist or data collection form) including validity and reliability  Describe who will collect the data  If you are going to use a (adaptation of a) standard tool, you can refer to an article in which it is described.  In case you developed a tool yourself, you should include it as an appendix. Designing a Questionnaire  Content ◦ Take your objectives and variables as your starting point ◦ Decide what questions will be needed to measure or define your variables and reach your objectives  Formulating questions ◦ Formulate one or more questions that will provide the information needed for each variable ◦ Check if each question measures one thing at a time ◦ Avoid leading questions ◦ Avoid words with double or vaguely defined meanings and emotionally laden words  Sequencing questions ◦ Design your questionnaire to be consumer friendly ◦ Must be sequential  Formatting the questionnaire ◦ Layout, heading, enough space to provide answers, boxes for answers are well placed  Translation  Data Analysis  This section should cover the variables to be measured and the statistical analysis to be done:  The statistical methods proposed to be used for the analysis of data should be clearly outlined including the program to be used.  The level of significance to be used should be stated.  How the data will be summarized (e.g. means, % including measures of variability e.g. SD, 95% CI). Ethical Consideration/Informed Consent  Make a line on ethics;  Will the study be sent for approval by an ethics committee;  Will (written) informed consent be given by participants? Havard Method of Referencing for articles  Author(s) (surname followed by initials) (year). Title of article. Name of journal, volume number: page numbers of article. After the first line, the other lines should be indented starting from the first name.  Examples for articles:  Eze CU, Eze CU, Adeyomoye A (2018). Sonographic evaluation of kidney echogenicity and morphology among HIV sero‑positive adults at Lagos University Teaching Hospital. Journal of Ultrasound, 21(1): 25-34.  Eze CU, Eze CU, Idogwu HA (2017). Evaluation of engagement in reflective practice by radiographers in Enugu metropolis, Southeast Nigeria. The South African Radiographer, 55 (1): 33 –38.  Eze CU, Ezugwu EE, Ohagwu CC (2017). Prevalence of cholelithiasis among Igbo adult subjects in Nnewi, Southeast Nigeria: a community-based sonographic study. Journal of Diagnostic Medical Sonography, 33 (2): 83-90.  Eze CU, Akpan, VP, Nwadike, IU (2016). Sonographic assessment of normal renal parenchymal and medullary pyramid thicknesses among children in Enugu, Southeast, Nigeria. Radiography, 22(1): 25 - 31. Vancouver Method: For articles  Author(s) (surname followed by initials). Title of article. Name of journal, year; volume number: page numbers of article.  Examples:  Luntsi G, Eze CU, Ahmadu MS, Audu AB, Ochie K. Sonographic evaluation of some abdominal organs in sickle cell disease patients in a tertiary health institution in Northeastern Nigeria. Journal of Medical Ultrasound, 2018; 26 (1): 29 - 34.  Abonyi EO, Eze CU, Onwuzu, SWI. Sonographic Correlation of Foetal Neck Circumference and Area with Gestational Age among pregnant women in Port Harcourt, Nigeria. Journal of Obstetrics and Gynaecology, 2017; 37 ( 8): 1025 - 1031.  Eze CU, Aguji EQ, Nwadike, IU. Sonographic Reference Values For Fetal Transverse Cerebellar Diameter in the Second and Third Trimesters in a Nigerian Population. Journal of Diagnostic Medical Sonography, 2017; 33 (3): 174-181.  Eze CU, Offordile GC, Agwuna KK, Ocheni S, Nwadike IU, Chukwu BF. Sonographic evaluation of the spleen among sickle cell disease patients in a Teaching Hospital in Nigeria. African Health Sciences, 2015; 15 (3): 949-958. Referencing Methods Contd  For a Book  Author(s) (surname followed by initials). Title of book. Edition. Place: Publisher, year: number pages in the book.  Example:  Abramson JH. Survey methods in community medicine. 2nd ed. Edinburgh: Churchill Living stone, 1979:229.  For a Chapter in a Book  Author(s) of chapter (surname followed by initials). Chapter title. In: Editors of book (surname followed by initials).. Title of book. Eds Place: Publisher, year: page numbers of chapter.  Example:  Winikoff B, Castle MA. The influence of maternal employment on infant feeding. In: Winikoff B, Castle MA, Laukaran VH, eds. Feeding infants in four societies: causes and consequences of mother choices. 2nd edition, New York: Greenwood Press, 1988: 12-145. Conclusion  THANK YOU ALL