1. THE TYPE OF RESEARCH DESIGN PRIMARILY IS BASED ON WHAT 12. SHOULD ALL POPULATION GET A SAMPLE?
ATION GET A SAMPLE? WHY OR WHY NOT?
ELEMENT OF RESEARCH? EXPLAIN o It depends. If the population is small, it can be the o Note that your research problem determines the type sample itself. However, population commonly contains of design you should use. too many individuals to study conveniently, so an o The type of research problem an organization is facing investigation is often restricted to one or more samples will determine the research design and not vice-versa. drawn from it. A well chosen sample will contain most o Variables, designated tools to gather information, how of the information about a particular population will the tools be used to collect and analyze data and parameter. other factors are decided in research design on the 13. GIVE THREE ELEMENTS OF A SAMPLE THAT WOULD SUPPORT basis of a research technique is decided. REPRESENTATIVENESS 2. CITE THREE DIFFERENCES BETWEEN EXPERIMENTAL AND NON o Shares a common attribute with the population EXPERIMENTAL RESEARCH DESIGN o Has a sound statistical sample (sample size) o Experimental o - has both testing and experiment 14. WHAT IS HEURISTICS AND WHEN DO WE USE IT? - there is an intervention in the treatment of the o Heuristics is the rule of thumb for the sample size used in variables the study, we use it if the population is infinite - all requires hypothesis 15. HOW DOES HETEROGENEITY OF POPULATION AFFECT SAMPLING? o Non experimental o If the population is homogenous, there are a lot of - has no experiment but has testing uniformity that exists among the population members a - no intervention in the treatment of variables sample of one will do. But if the population is - does not need/require hypothesis heterogeneous, a sampling technique that will widely 3. HOW IS RESEARCH CONTROL UNDERTAKEN IN EXPERIMENTAL spread the choosing of a large sample among all RESEARCH DESIGN? members of the population is necessary o In experimental research design, the research control is 16. WHY IS SYSTEMATIC SAMPLING EASIER DONE THAN SIMPLE the group that is not given any treatment or condition RANDOM SAMPLING? 4. GIVE A SCENARIO WHEN THE RESEARCHER CAN NOT RANDOMIZE o Because systematic sampling is more organized than HIS RESEARCH SUBJECTS. random sampling o If the samples are limited and there is an inclusion criteria 17. DIFFERENTIATE STRATIFIED SAMPLING FROM CLUSTER SAMPLING. GIVE 2 5. WHAT’S THE IMPORTANCE OF PRETESTING? GIVE TWO TANGIBLE o Stratified – choosing a sample that will be subdivided REASONS. into strata or sub-groups | homogenous o To identify if there are some error in order to eliminate it o Cluster – selecting respondents from a cluster | o To know if there is a change in the DV homo/hetero grouping o To check the effectivity of intervention 18. DIFFERENTIATE QUOTA FROM PURPOSIVE SAMPLING. GIVE 2 6. WHY IS SOLOMON FOUR GROUP DESIGN BETTER THAN PRETEST- o Quota – inclusion criteria is lax POST TEST DESIGN? o Researchers employ a pretest-posttest design in order o Purposive – inclusion criteria is rigid to demonstrate that exposure to a treatment led to differences between the pretest and posttest scores. 19. DIFFERENTIATE AVAILABILITY FROM SNOWBALL SAMPLING. GIVE 2 However, there can be some drawbacks to including o Availability – researchers seeks for the respondents with both a pretest and posttest in an experimental design. consent | convenience/ambush sampling In particular, researchers have noted that including a pretest in a study design introduces threats to both o Snowball – network sampling | good if sample size is internal and external validity. The Solomon four group very rare (referral) test is a standard pretest-posttest two-group design and the posttest only control design. The various 20. WHEN IS SNOWBALL SAMPLING BEST EMPLOYED? combinations of tested and untested groups with o If the sample size is very rare (referral) treatment and control groups allows the researcher to ensure that confounding variables and extraneous factors have not influenced the results, thus, it has an 21. GIVE 2 DISADVANTAGES OF CLUSTER SAMPLING advantages of strengthening internal and external o Prone to bias when the number is small validity o High sampling error 7. ALL RESEARCH STUDIES WITH HYPOTHESIS EMPLOY EXPERIMENTAL DESIGN. AGREE OR DISAGREE? WHY 22. DIFFERENTIATE QUANTITATIVE DATA COLLECTION FROM o Agree, because all experimental research design QUALITATIVE DATA COLLECTION. GIVE 2 requires hypothesis and non experimental research o design, does not require hypothesis 23. HOW CAN YOU CLASSIFY AN INSTRUMENT TO BE ADAPTED? GIVE AT LEAST 2 CRITERIA 8. WHICH HAS BETTER PREDICTABILITY, COMPARATIVE OR o CORRELATIVE STUDIES? WHY? o Correlative studies have better predictability since o if there is a relationship between two variables, we can make predictions about one from another. 24. WHAT SPECIFIC SURVEY TECHNIQUE HAS HIGHER RETURN RATE 9. EXPLAIN WHY EX POST FACTO IS CONSIDERED A NON AND LOWER MORTALITY? EXPLAIN EXPERIMENTAL DESIGN o Interview o Because it derives data and explanations from past o Higher return rate – ask a set of questions to a certain events, it determines the cause and effect but has no respondents, orally treatment or intervention o Lower mortality – takes time in doing this 10. GIVE 3 FEATURES OF A QUANTITATIVE RESEARCH DESIGN 25. WHAT IS A TIME SERIES EXPERIMENT? GIVE A SPECIFIC EXAMPLE o It has to types: experimental and non experimental o o It focuses on the methods to be used in looking for answers in your study 26. HOW DO YOU DO A QUANTITATIVE CONTENT ANALYSIS? o It serves as the bloodline of a research study o 11. BEFORE DECIDING WHAT RESEARCH DESIGN TO USE WHAT WILL BE YOUR BASIS? GIVE 3 27. WHICH HAS STRONGER DATA VALIDITY? DISCRETE OR o Methods/techniques in finding answers to your CONTINUOUS VARIABLE? EXPLAIN research questions and in collecting data o The one with the stronger data validity is continuous o variable since the data that will be obtained is more o precise and it can be measured, unlike in discrete o Instrumentation is the process of creating a tool variables that it is only categorized. 41. GIVE THE THREE STATISTICAL QUALIFICATION OF A PARAMETRIC 28. DIFFERENTIATE VALIDITY FROM RELIABILITY. GIVE TWO EXAMPLES TEST FOR EACH o Data – interval or ratio (continuous) o Validity – extent to which the instrument measures o Size – 30 or more what it intends to measure (Face and Content Validity) o Normality – data is normal o Reliability – relates to the extent which the instrument is 42. DIFFERENTIATE PARAMETRIC AND NON PARAMETRIC TESTS consistent (Equivalence, Inter-rate reliability) o Parametric – number of assumptions | has high 29. WHAT IS INTERNAL CONSISTENCY? EXPLAIN THE THREE WAYS OF assumptions on data | has higher statistical power MEASURING IT (generalization) o Internal consistency – the extent to all items on a scale o Non Parametric – few or no assumptions | has low measure one construct statistical power - Split half coefficient is obtained by 43. DIFFERENTIATE T-TEST FOR DEPENDENT AND T-TEST FOR administering a single instrument INDEPENDENT MEANS - Cronbach’s alpha measures the reliability with o Dependent – paired t test (1 grp) respect to each item & construct being o Independent – unpaired t-test (2 grps) examined 44. WHAT IS BIVARIATE ANALYSIS? GIVE TWO EXAMPLES - Kuder Richardson Formula tests the reliability in o Bivariate is a type of statistical data analysis that terms of instruments of a dichotomous nature analyses 2 variables (ex: yes or no) o T-test, one way ANOVA, Pearson’s r 30. HOW IS STABILITY RELIABILITY DONE? WHAT STATISTICAL TOOL IS 45. WHAT IS A MULTIVARIATE ANSLYSIS? GIVE TWO SPECIFIC USED? EXAMPLES o o A type of statistical data analysis that analyses multiple relations between multiple variables 31. WHEN IS KAPPA COEFFICIENT UNDERTAKEN? HOW IS IT DONE? o Multiple regression, MANCOVA o Kappa coefficient – one of the most popular statistical 46. EXPLAIN THIS CORRELATION COEEFICIENT -0.39 tool in measuring inter-rater reliability o Negative, weak o It is done through 47. DIFFERENTIATE POSITIVE FROM NEGATIVE CORRELATION o Positive – when the numerical value of one variable 32. DIFFERENTIATE CONCURRENT VALIDITY FROM PREDICTIVE increases or decreases, the other variable increases or VALIDITY. GIVE TWO decreases as well (directly proportional) o Both – under criterion validity o Negative – as the numerical value of one variable o Concurrent – it is able to predict results similar to those increases, the other variable decreases, and vice-versa of a test already validated in the past (indirectly proportional) o Predictive – produces results similar to those of another 48. HOW DOES THE P VALUE AND ALPHA LEVEL AFFECT EACH OTHER? instrument that will be employed in the future EXPLAIN 33. DIFFERENTIATE CONSTRUCT FROM CONTENT VALIDITY. GIVE TWO o The p-value measures the probability of getting a more o Construct – degree to which an instrument measures extreme value than the one you got from the the variables being studied as a whole experiment. If the p-value is greater than alpha, you o Content - coverage of a tool | refers to the degree to accept the null hypothesis. If it is less than alpha, you which an instrument covers a representative sample of reject the null hypothesis. variables to be measured 49. DIFFERENTIATE ONE WAY ANOVA FROM ANCOVA 34. WHAT IS CODING SYSTEM? WHEN DO YOU DO THIS? o One way ANOVA – test differences among groups o Assigning numerical digit for non numerical data | 1 IV – factor, 1 DV – response changing verbally expressed data into numerical info o ANCOVA – study of 2 or more DV that are correlated o we are doing this when our variables are non numerical with each other like gender 2 IV (factor & covariate), 1 DV (response) 35. DIFFERENTIATE KURTOSIS AND SKEWNESS. EXPLAIN THEIR ROLE IN 50. GIVE THREE TECHNICAL TIPS IN WRITING YOUR RESEARCH REPORT NORMALITY o Organize the parts of your research report based on o Kurtosis – measure of the heaviness of the tail of the the standard research-report structure that consists of distribution, thus indicating the number of outliers in the following your data sequential components: (Title, abstract, o Skewness – the lack of evenness in the distribution of introduction,methodology, results and findings, data conclusions,recommendation, references and 36. TABULATION HAS TWO FUNCTIONS. WHAT ARE THESE? EXPLAIN. appendices) o Data preparation o Familiarize yourself with the language of academic writing. (Use passive than active, used third POV, use o Data analysis modality) o Observe the mechanics of research-report writing 37. WHY IS RANGE NOT A DEFINITIVE MEASURE OF DISPERSION? (PA, Quotations, footnote, EXPLAIN. statisticd and graphs, final draft, index) o The range is the difference between the largest and the smallest observation in the data. The prime advantage of this measure of dispersion is that it is easy to calculate. On the other hand, it has lot of disadvantages. It is very sensitive to outliers and does not use all the observations in a data set. It is more informative to provide the minimum and the maximum values rather than providing the range. 38. DIFFERENTIATE MEAN FROM STANDARD DEVIATION o Mean – average of all items and scores o Standard Deviation – shows the extent of the difference of the data from the mean 39. DIFFERENTIATE DESCRIPTIVE FROM INFERENTIAL STATISTICS o Both – approaches/statistical methodologies o Descriptive – describes certain aspect of a data | has no hypothesis o Inferential – leads to generalization/conclusions in the population | has hypothesis testing 40. EXPLAIN BRIEFLY THE PROCESS OF INSTRUMENTATION