Research Methods 1 Notes PDF

SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ୨ৎSession 1 discussion: ★Better to distinguish assumptions of each method and the kinds of data that are collected....

SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ୨ৎSession 1 discussion: ★Better to distinguish assumptions of each method and the kinds of data that are collected. PHILOSOPHY ★Variable (Hypothesis) - characteristics of what you study. ★Putting forward by using Hypothesis / Assumptions Correlation Matrix ★Essence - Ontology ★Needs - Utility ★Existence - Existential ★Theme 1 correlation with Theme 2 and so on. ★Person 1 correlation with Person 2 and so on. Settling the Debate 1. Qualitative data can be coded quantitatively. 2. All quantitative data is based on qualitative judgment. 3. Content analysis - archival documents, second data analysis, popular media, meta-analysis. ★Make clear of the concepts you will put in your questionnaires Durkheim’s: to avoid confusion. ★Avoid redundant words in your sentence when making questionnaires. ASSUMPTIONS ★The heart of quantitative-qualitative debate is Philosophical, not methodological. Positive (Durkheim) Interpretative (Gabriel Tarde) ★Different epistemological assumptions (POST) Positivism vs. Constructionism. ★Social Facts ★Monads ★Different ontological assumptions about the world. - Created by the - Individual alone can interaction between two order the world TWO COMMON MYTHS: or more people, which is called COLLECTIVE CONSCIENCE ★Quantitative research is confirmatory and deductive in nature. ★Qualitative research is exploratory and inductive in nature. ★Objective ★Subjective ★Inductive (Qualitative) - Starts with a very specific question but ends up with a Wilhelm Windelband very general conclusion. - Also called as Bottom-up approach (open-ended) Nomothetic (Quantitative) Idiographic (Qualitative) - Examines pattern and irregularities - Theory generating ★Generalized, majority ★Specific, unique, particular ★Deductive (Quantitative) - start with a very general question but ends up with a very specific conclusion. The Quali-Quanti Debate - Also called as Top-down approach (narrow) - Assumptions of theory (operationalized) ★Qualitative surveys then forward to Quantitative - Theory conforming and debunking questionnaires. - Mix Methods Research (Quali-Quanti) QUALITATIVE DATA ★Much larger, much richer Are mainly gathered through: PAJOILAS, HAZEL ANNE S. | 1 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★Surveys (POST) Sociological: their experience on why they don’t want to have - Dependent and independent variables will study kids. retrospectively. - Retrospective and representative ★Humans are prone to retroactivity, which means the margin of error is higher. ★Experiments (PRE-POST INTERACTION) - Manipulate or eliminate some variables 1st: Ceteris Paribus - all things are held equal or constant. - Association and direction 2nd: Suspend or manipulate the variables (Economists), - POSITIVE - directly experiments (Psychologists) - NEGATIVE - inversely proportional - Mas mataas ang internal validity ★Meta-analysis ★Content Analysis - You manipulate the result. - You allow others to criticize it. ★Participation observation ★Participant Observation - Participating while observing. ANALYSIS OF QUANTITATIVE DATA CAN BE: ★Direct - aware that they are being observed ★UNIVARIATE (Simple Tabulate) - the retro-active effect but you can see the changes - Measures of central tendency and variability. ★Indirect - Mean (ave), median (middle), mode (most often - they don’t know that they are being observed occurred) [Median: 2nd Quartile] - Effect: Loss of Ethics ★Outlier (extreme high, extreme low) - Average score is affected Formula of Percentage: ★if the mean and median is equal, it is normal % = observe number of case / total number of case x 100 ★if the mean is lesser than median, it is left ★if the mean is greater than median, it is right QUALITATIVE APPROACHES ★BIVARIATE - Parametric and non-parametric test. ★ Is a general way of thinking about conducting qualitative - Interval ratio (Parametric Test is for Psychologists) research. It describes, either explicitly or implicitly, the purpose - Non-Parametric Test is for Sociologist (Ordinal) of the qualitative research, the role of researcher(s), the stages of - From ordinal to interval and interval to ordinal research, and the methods of data analysis. ★MULTIVARIATE - Interaction between dependent variable to independent FOUR MAJOR QUALITATIVE APPROACHES variable [Variability - gaano kalapit, gaano ka-spread out] ★Ethnography / Anthropology - Require investment of time - Particular animate the general - Persistence of the past to present ୨ৎSession 2 discussion: ★Phenomenology - Affect each others [REBT - no social forces] QUANTITATIVE DATA ★Rational ★Synthesis the RRL ★Emotive (In sociology, you need to see the ★Quantitative (the variables are clear and established); ★Behavior connection to social forces) ★Qualitative (Unique cases) ★Therapy ★Field Research “How social forces affect the individual’s action” - Sometimes doesn’t include ethnography ★Survey question is different from the instrument question. - Not all has a lot of investment of time unlike ethnography ★DINKS - Double Income No Kids PAJOILAS, HAZEL ANNE S. | 2 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★Grounded Theory ★Conclusion - Approach not a method - Relationship between variables - Requires open-mindedness - Null case – treatment has no effect - Suspend all theories/lenses, have an open mind not ★Internal empty mind. - Tama ba yung ginamit na design ★Construct QUALITATIVE DATA - Nagpi-fit ba yung questionnaire - Nakakapag-operationalized ba siya ★In-depth interviews – participants / interlocutors - Kapag mali, affected ang instruments ★Epistemology ★External - No structured questions - Sampling Methods - Has guide questions ★Direct Observation FURTHER ELABORATION ★Written Documents ★Participant Observation ★Credibility ★Unstructured Interviewing - Will the participants of the research find your findings believable? Will they endorse your result? Must they ★Case Studies also agree or support your theory? [Read: Living the Priesthood] ★Transferability - Can other researchers apply your results (your theory) QUALITATIVE VALIDITY in other settings? E.g. Nature of Leadership in Whyte’s Street Corner Society, or the loss of idealism in Becker ★Some qualitative researchers reject the framework of validity and Geer’s Boys in white. that is commonly accepted in more quantitative research in the ★Dependability social sciences. - The qualitative version of reliability testing. How dependable are your results? Depends on what ★Rejection of the realist assumption that there is a reality context? What if the context changes? external to our perception of it. ★Confirmability - To what extent can your results be confirmed or ★Theoretical Sampling – satisfied your criteria. corroborated by others – not only by the research participants themselves. Traditional criteria for judging Alternative criteria for judging quantitative research qualitative research ★Dependent - Nominal (mahirap makita ang direction) Internal Validity Credibility ★Dependent and independent - Ordinal (mabilis makita ang direction) External Validity Transferability Reliability Dependability Objectivity Confirmability ୨ৎSession 3 discussion: ★External validity – generalization Read: Elements of Style by Strunk and White ★Reliability – instrument; consistency YIN YANG OF QUANTITATIVE RESEARCH ★Validity – findings ★Confirmability – citation RESEARCH Theory Research practice FOUR TYPES OF VALIDITY Construct Validity Measurement PAJOILAS, HAZEL ANNE S. | 3 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 Internal Validity Design ★Example - A public opinion poll that compares the proportion of Conclusion Analysis males and females that say they would vote for a Democratic or a Republican candidate in the next External Validity Sampling presidential election is essentially studying the relationship between gender and voting preference. ★Causal studies (Experimental studies) [Pre and Post] RESEARCH VOCABULARY - Are designed to determine whether one or more variables (for example, a program or treatment variable) cause or affect one or more outcome ★Theoretical variables. - Pertaining to theory. Social research is theoretical, - Multi group and Multivariate meaning that much of it is concerned with developing, - Eliminate third variable problem (spurious – mirage), exploring, or testing the theories or ideas that social eliminating by assigning a control variable. researchers have about how the world operates. ★Example ★Empirical - If you performed a public opinion poll to try to - Based on direct observations and measurement of determine whether a recent political advertising reality. campaign changed voter preferences, you would ★Probabilistic essentially be studying whether the campaign (cause) - Based on probabilities. changed the proportion of voters who would vote ★Causal Democratic or Republican (effect). - Pertaining to a cause-effect relationship - Experiments TIME DIMENSION IN SOCIAL RESEARCH - Purify the IV and DV to analyze the control variable ★Correlation and Elective Affinity ★Cross Sectional - Elective Affinity (Protestant Ethics and Spirit of - They examine a single point in time or take a one-time Capitalism by Weber, the 2 historical events) snapshot approach - Korrelation vs. Wahlverwandtschaft (valfervanschaft) - In effect, you are taking a slice or cross-section of - Correlation: relationship; identified ang IV at DV whatever it is you’re observing or measuring. - Elective Affinity: hindi identified yung IV at DV - E.g. PISA ★Placebo Effect ★Longitudinal - Akala mo effective rin sayo kasi effective sa iba. - Study that takes place over time - E.g. Homeopati - believing that it will work - You measure your research participants on at least two - E.g. Homeostasis - capacity of the body can repair itself separate occasions or at least two points in time. - When you measure at different time points, we often TYPES OF STUDIES say that you are measuring multiple waves of measurement. ★Descriptive studies (Univariate) - Are designed primarily to document what is going on ★Time Series or what exists. - Observe different people at multiple times - Public opinion polls that seek to describe the proportion - Examining the innovation or changes from past to of people who hold various opinions are primarily present. descriptive in nature. - E.g. 1950, 1970, 1990, 2010 ★Example ★Panel - If you want to know what percent of the population - Observe the exact same people at two or more times would vote for a Democrat or a Republican in the next - E.g. 1986, 1996, 2006 presidential election, you are simply interested in ★Cohort describing something. - Observe people who shared similar life experience at ★Relational studies (Bivariate) two or more time - Relationships between two or more variables. - E.g. Married in 1967, 1987, 2007 PAJOILAS, HAZEL ANNE S. | 4 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★Case study - For example, when you evaluate whether your treatment - Observe a small set intensely across time or program causes an outcome to occur, you are - E.g. 2005 to 2007 examining a causal relationship. ★Third variable problem - An unobserved variable that accounts for a correlation VARIABLES between two variables. Variation among people ★Attribute (determining the scale of your variable) PATTERNS OF RELATIONSHIPS (DIRECTIONALITY) - A specific value of a variable. For instance, the variable sex or gender has two attributes: male and female. - The variable agreement might be defined as having five ★Positive relationship (directly proportional) attributes: - A relationship between variables in which high values ★1= strongly disagree for one variable are associated with high values on another variable, and low values are associated with low ★2= disagree values. ★3= neutral - E.g. Law of supply, when the price increases, the supply ★4= agree also increases. ★5= strongly agree ★Negative relationship (indirect proportional) - A relationship between variables in which high values TYPES OF VARIABLES for one variable are associated with low values on another variable. - E.g. Law of demand, when the price increases, the ★Independent variable (presumed caused) demand decreases. - The variable that you manipulate. - For instance, a program or treatment is typically an a.) No relationship b.) A positive relationship independent variable ★Dependent variable (presumed outcome) - The variable affected by the independent variable - For instance, the outcome CONDITIONS OF ATTRIBUTES ★Exhaustive (Researchers) - The property of a variable that occurs when you include all possible answerable responses. c.) A negative relationship d.) A curvilinear relationship - Indication of others – include all possible answers ★Mutually Exclusive (Respondents) - The property of a variable that ensures that the respondent is not able to assign two attributes simultaneously - For example, gender is a variable with mutually exclusive options if it is impossible for the respondents to simultaneously claim to be both male a X= Abscissa (independent) TYPES OF RELATIONSHIP Y= Ordinate (dependent) y= mx+b ★Relationship - Refers to the correspondence between two variables. ★Curvilinear relationship has a third variable and it is mainly used in ★Causal relationship natural science, especially in medicine. - A cause-effect relationship. PAJOILAS, HAZEL ANNE S. | 5 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 Where to find research literature? HYPOTHESES ★People ★Periodicals ★Hypothesis ★The internet - A specific statement of prediction ★Books ★Alternative hypothesis ★Dissertation - A specific statement of prediction that usually states ★Policy Reports what you expect will happen in your study. ★Government Documents - There is a significant relationship between IV and DV. ★Null hypothesis THREE MAIN DOCUMENTATION PROCESS - The hypothesis that describes the possible outcomes other than the alternative hypothesis. Usually, the null hypothesis predicts there will be no effect of a program ★People or treatment you are studying. - Use the network method to conduct this first stage of - You reject the relationship between IV and DV your literature review. ★Review articles and bibliographies ୨ৎ1st Reporting Topic: - The Annual Review series is a good place to start reading. There are Annual Review volumes for many disciplines, including psychology. STRATEGIES IN COLLECTING AND ANALYZING ★A host of online databases LITERATURE AND STUDIES - Bibliographic Search tools ★Literature ★Web of Science - Collection of studies already published on a topic - Social Sciences CItation Index (the SSCI) - It helps narrow down a broad topic by showing you how - Arts and Humanities Citation Index others conducted their studies. - “State of knowledge” ★Other documentation databases - Stimulates creativity - Anthropological Index Online (AIO) - improves our own skills - Anthropological Literature (AL) - shows the final look of a published research study - Abstracts in Anthropology ★Review of Related Literature - Is based on the assumption that knowledge ★Other useful databases include: accumulates and that people learn from and build on - International Bibliography of Anthropology what others have done. - ERIC - NTIS GOALS OF LITERATURE REVIEW - MEDLINE - PsycINFO 1. Demonstrate a familiarity with a body of knowledge and - Sociological Abstracts establish credibility. - Lexis-Nexis 2. To show the path of prior research. - OCLC 3. Integrate and summarize what is known in an area. - JSTOR 4. To learn from others and stimulate new ideas. - EBSCO HOST - IBSS ★Source file - CORE - Record of all the bibliographic information for each source. ★Predatory journals ★Content file - Review process that doesn’t go on the standard review - Substantive information of interest from a source. evaluation. - beallslit.net - predatoryjournals.org PAJOILAS, HAZEL ANNE S. | 6 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ୨ৎSession 4 Discussion: reduce depression. (After all, you’ve seen more than enough promising drug treatments come along that eventually were shown to have severe side effects that RESEARCH VOCABULARY CONTINUATION actually worsened symptoms.) HYPOTHESES ★The null hypothesis in this study is: ★ For instance, let’s imagine that you are investigating the ★HO: As a result of 300mg/day of the ABS drug, there will be effects of a new employee-training program and that you believe no significant difference in depression, one of the outcomes will be that there will be less employee Which is tested against the alternative hypothesis: absenteeism. Your two hypotheses might be stated something like ★HA: As a result of 300mg/day of the ABC drug, there will be a this: significant difference in depression. ★HO: As a result of the XYZ company employee-training ★The term two-tailed refers to the tails of the distribution for program, there will either be no significant difference in your outcome variable. employee absenteeism or there will be a significant increase. ★HA: As a result of the XYZ company employee-training program, there will be a significant decrease in employee Nota bene: absenteeism. ★The important thing to remember about stating hypotheses ★Null case is that you formulate your prediction (directional or not), and then you formulate a second hypothesis that is mutually - A situation in which the treatment has no effect. exclusive of the first and incorporates all possible alternative outcomes for that case. ★One-tailed hypothesis - A hypothesis that specifies a direction; for example, ★When your study analysis is completed, the idea is that you when your hypothesis predicts that your program will will have to choose between the two hypotheses. increase the outcome. - A value of zero means that there has been no difference in absenteeism observed, a positive value ★If your prediction is correct, you would (usually) reject the means that absenteeism has increased, and a negative null hypothesis and accept the alternative. value means it has decreased. - The term one-tailed refers to the tail of the distribution ★If your original prediction was not supported in the data, on the outcome variable. you will accept the null hypothesis and reject the alternative. - Implies direction HYPOTHESES IN GENERAL ★Two-tailed hypothesis - A hypothesis that does not specify a direction ★Two mutually exclusive hypothesis statements that, together, - For example, if your hypothesis is that your program or intervention will have an effect on an outcome, but you exhaust all possible outcomes need to be developed. are unwilling to specify whether that effect will be ★The hypotheses must be tested so that one is necessarily positive or negative, you are using a two-tailed accepted and the other rejected. hypothesis. - Doesn’t implies direction, just association ★Hypothetico-deductive model - Survey (for open-ended) - A model in which two mutually exclusive hypotheses that - Present null and alternative hypothesis. together exhaust all possible outcomes are tested, such that if one hypothesis is accepted, the second must ★As a case in point: therefore be rejected. - The drug has gone through some initial animal trials, ★Ideology but it has not yet been tested on humans. You believe - Hindi na kayang i-debunk (based on theory and the previous research) that the ★Dogma drug will have an effect, but you are not confident - Religious (cannot be falsified) enough to hypothesize a direction and say the drug will PAJOILAS, HAZEL ANNE S. | 7 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★Argumentum ad misericordiam TYPES OF DATA - Appeal to pity (not rational thinking) ★Quantitative data ★Ad Baculum - Data that appears in numerical form - Appeal to force ★Qualitative data (descriptive) - Data in which the variables are not in a numerical form, STRUCTURE OF RESEARCH but are in the form of text, photographs, sound bytes, and so on. The hourglass metaphor for the research process (thesis, antithesis, synthesis) UNIT OF ANALYSIS ★Both ends of the hourglass The entity that you are analyzing in your analysis; for example, represent the realm of ideas individuals, groups, or social interactions. and the research questions that guide the project. ★Individuals ★The hourglass center is the ★Groups most concrete or specific part of ★Artifacts (books, photos, newspapers) the process. ★Geographical units (town, census, tract, state) ★The parts in between show how we translate the research ★Sociological interactions (dyadic relations, divorces, arrests) questions into procedures for measurement (top part of the ★Scalar Theoretical Disconnect hourglass) and how we translate the data we observe into - Hindi nag fit yung theory sa topic mo conclusions and new or revised questions (bottom part). ★In the operationalize section, it comes with indices, to what extent, IV, DV, and in other terms, it is to actualize. ୨ৎSession 5 Discussion: COMPONENTS OF A STUDY THE RATIONALE OF RESEARCH ★Most social research originates from some general research ★A research project has a well-known structure: a beginning, problem. middle, and end. ★Usually, the problem is broad enough that you could not hope ★Before the modern idea of research emerged, there was a to address it adequately in a single research study. term for what philosophers used to call research: Logical ★The problem is typically narrowed down to a more specific reasoning. research question that can be addressed. ★How two major logical systems—the inductive and deductive methods of reasoning—are related to modern research. ★Research question An error in logic is often called a fallacy, and there are a - The central issue being addressed in the study, which is number of fallacies that tend to come up repeatedly in typically phrased in the language of theory. The conducting research. problem with such a question is that it is all too general ★Issues of Validity: How do we know whether the answers we to be studied directly. get from research are sensible and accurate? How valid are the results of your research? What factors contribute to making a ★Consequently, in much research, an even more specific study more or less valid, stronger or weaker? statement, called a hypothesis, is developed that describes in operational terms exactly what you think will happen in the study. ★Logic – fallacy (a pitfall for your logical thinking. ★Ad Hominem ★Notice that the hypothesis is specific enough that a reader - Attack the person, nit the work of person can understand quite well what the study is trying to assess. PAJOILAS, HAZEL ANNE S. | 8 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★In causal studies, there are at least two major variables of problem; usually, the greatest interest is in outcomes that are interest: the cause and the effect. most reflective of the problem. ★A distinction is made between causes that the researcher can ★In a causal study, the effects of the cause of interest (for control (such as a program) versus causes that occur naturally example, the program) are usually compared to other conditions or outside the researcher’s influence (such as a change in (for example, another program or no program at all). interest rates, or the occurrence of an earthquake). ★Thus, a key component in a causal study concerns how you ★The effect (or dependent variable) is the outcome that you decide which units (people) receive the program and which are wish to study. placed in an alternative condition. ★This issue is directly related to the research design that you ★Journalists use in the study. - Doesn’t formulate a theory ★One of the central themes in research design is determining ★Sociologists how people wind up in or are placed in various programs or - Produces theoretical knowledge treatments that you are comparing. - Empirically driven, theoretically oriented DEDUCTION AND INDUCTION ★Dissertation or law is a dependent variable, for instance, law of supply and law of demand. ★In logic, a distinction is often made between two broad methods of reasoning known as the deductive and inductive ★For both the cause and effect, a distinction is made between approaches. the idea of the cause or effect (the construct) and how they are actually manifested in reality. ★Deductive reasoning In research, a distinction is made between your view of works from the more an entity (the construct) versus the entity as it exists general to the more (the operationalization). specific ★Sometimes, this is ★Idea (construct) – cause-effect informally called a ★Construct – your view [how you theoretically think of the entity top-down approach (variable)] ★This ultimately leads ★Operationalized – entity exists as itself [to translate your you to be able to test the hypotheses with specific data—a questionnaires into reality] confirmation (or not) of your original theories. ★It is narrower in nature and is concerned with testing or ★Ideally, the two should agree, but in most situations, reality falls confirming hypotheses. short of your ideal. ★Inductive reasoning ★Issues of Units and Sampling works the other way, - Social research is always conducted in a social context. moving from specific Researchers ask people questions, observe families observations to broader interacting, or measure the opinions of people in a city. generalizations and ★EPSEM theories. - Equal Probability of Selection Method ★In formally, this is - Problem with external validity, hindi mage-generalized sometimes called a bottom-up approach (threats to validity) ★You begin with specific observations and measures, begin - If the construct and sampling is incorrect, you cannot detecting patterns and regularities, formulate some tentative also achieved the EPSEM hypotheses that you can explore, and finally end up developing some general conclusions or theories. CAUSAL STUDIES ★By its nature, it is more open-ended and explanatory, especially at the beginning. ★The interest is in the effects of some cause on one or more outcomes. The outcomes are directly related to the research PAJOILAS, HAZEL ANNE S. | 9 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 Remember the Acronym CICE NOTA BENE: Even though a particular study may look like it’s purely ★Conclusion Validity deductive (for example, an experiment designed to test the - In this study, is there a relationship between the two hypothesized effects of some treatment on some outcome), variables? most social research involves the inductive and deductive ★Internal Validity reasoning processes at some time in the project. - Assuming that there is a relationship in this study, is the relationship a causal one? TWO RESEARCH FALLACIES ★Construct Validity - Did you implement the program you intended to ★Ecological Fallacy implement, and did you measure the outcome you - Faulty reasoning that results from making conclusions wanted to measure? about individuals based only on analyses of group data. - Did you operationalize the ideas of the cause and the - True in general, but not always true in particular effect well? ★Exception Fallacy ★External Validity - A faulty conclusion reached as a result of basing a - Assuming that there is a causal relationship in this conclusion on exceptional or unique cases study between the constructs of the cause and the - True in particular but not always true in general effect, can you generalize this effect to other persons, places, or times? VALIDITY OF RESEARCH ★Threats to validity - Reasons your conclusion or inference might be wrong. ETHICS IN RESEARCH (VICAR) V – VOLUNTARY participation Do not coerced I – INFORMED CONSENT Consent to participate C – CONFIDENTIALITY Assurance to not be released to anyone A – ANONYMITY No one will be able to link data to anyone R – RIGHT TO SERVICE Everyone has the right to receive it ★Voluntary participation - For ethical reasons, researchers must ensure that study participants are taking part in a study voluntarily and ★Validity are not coerced. - The best available approximation of the truth of a given ★Informed consent proposition, inference, or conclusion. - A policy of informing study participants about the ★Cause construct procedures and risks involved in research that ensures that all participants must give their consent to - Your abstract idea or theory of what the cause is in a participate. cause-effect relationship you are investigating. ★Effect construct ★Confidentiality - An assurance made to study participants that - Your abstract idea or theory of what the outcome is in identifying information about them acquired through a cause-effect relationship you are investigating. the study will not be released to anyone outside of the study. PAJOILAS, HAZEL ANNE S. | 10 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★Anonymity ୨ৎSession 6 Discussion: - The assurance that no one, including the researchers, will be able to link data to a specific individual. ★Statistics ★Right to service - “A set of procedures for gathering, measuring, - The ethical issue involved when participants do not classifying, coding, computing, analyzing, and receive a service that they would be eligible for if they summarizing systematically acquired numerical were not in your study. For example, members of a information (Ritchey 2008:1).” control group might not receive a drug because they are in a study. ★Probability (Correlation) - Compare and contrast from the general population INSTITUTIONAL REVIEW BOARD (IRB) ★Descriptive ★A panel of people who review research proposals with respect - Univariate, monovariable to ethical implications and decide whether additional actions - Just describing your data need to be taken to assure the safety and rights of participants. ★Inferential - Bivariate, multivariate CONCEPTUALIZING - You’re making an interpretation STATISTICS AND SCIENCE S M A R T Specific measurable attainable relevant time ★Science is a systematic method of explaining empirical Plan Track your Set realistic Ensure the Specify a phenomenon. effectively progress goals that goal serves deadline, with and are a relevant monitor - Empirical phenomena are things that can be observed specific reevaluate challenging purpose. progress or measured such as natural conditions, processes, targets in along the but and events, situations, objects, groups of people, behavior, mind. way. achievable. reevaluate. thoughts, beliefs, etc. ★Phenomena WHAT SHOULD YOU LOOK FOR IN THE LITERATURE - Can be measured REVIEW? - Science (being empirical), it is falsifiable and creates theories Its main objective is to explain things. ★First, you might be able to find a study that is quite similar to ★Theory the one you are thinking of doing. Since all credible research - A set of interrelated, logically organized statements that studies have to review the literature themselves, you can check explain a phenomenon of special interest and that have their literature review to get a quick start on your own. been corroborated through observation and analysis. - Temporary analysis (it could change over time), ★Second, prior research will help ensure that you include all of remember that science is falsifiable the major relevant constructs in your study. You may find that Ex. Durkheim’s Theory of Suicide other similar studies routinely look at an outcome that you might - Mainly those single protestant male are prone to not have included. Your study would not be judged credible if it suicide during his time, but it can change over time, ignored a major construct. generation by generation. ★You always dare to question science. ★Third, the literature review will help you find and select ★Science requires that its ideas withstand the test of being able appropriate measurement instruments. You will readily see what to predict observations. measurement instruments researchers used themselves in a - Falsifiability as trade practice. Skepticism as attitude. context similar to yours. ★Science has its limitations ★Finally, the literature review will help you to anticipate common - Focus on the empirical closes off the non-empirical (the subjective part cannot be measured) problems in your research context. You can use the prior - Many sound arguments lack political or taxpayer experiences of others to avoid common traps and pitfalls. support PAJOILAS, HAZEL ANNE S. | 11 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 - Can create ethical dilemmas in application ★Statistical analysis is a key feature of the scientific method. PROPORTIONAL THINKING - Seeks precise understanding and control of statistical error (gaano kalapit sa population mo) ★Weighing the part against the whole and cultivating the likelihood of the phenomenon occurring over the long run. ★Census - Describing the entire population ★Survey Fractions Find the Greatest Common Divisor (GCD) - Describing the sample population Proportions Quotient of the fractional form (decimal) ★Confidence interval - Mataas ang confident mo na roon magfafall Percentages Proportion x 100 - range of values you expect your estimate to fall between if you redo your test, within a certain level of Rates Frequency of occurrence (larger than confidence. hundreds—thousands, millions) - ±5 [interval score] Rate of occurrence = (p) (a useful base number) ★Example - 67 : 67-5= 62 Rates to Ratio Rate / base number = 1/x people - 67: 67+5= 72 X = (base number/rate) - It can fall between 62 to 72 1.8/10,000,000 = 1/x people X = 10,000,000/1.8 = 5,555.556 people ★Proportions - Part of the total amount or number of observations, expressed in decimal form. The quotient of fraction. Example ★48/60 = 4:5 or.8 ★45/75 = 3:5 or.60 ★84/126 = 2:3 or.67 ★ 144/216 = 2:3 or.67 ★56/98 = 4:7 or.57 ★63/189 = 1:3 or.33 ★Percentage - The proportion multiplied by 100 FRACTIONS ★Mathematical proportions are simply division problems that weigh the part (numerator) against the whole (denominator). To compute a proportion, we start with a fraction. FRACTION= NUMERATOR (PART) / DENOMINATOR (WHOLE) Example Examples: ★4/5 = 80% ★3/5 = 60% ★48/60 = ⅘ ★45/75 = ⅗ ★2/3 = 67% ★2/3 = 67% ★84/126 = ⅔ ★ 144/216 = 2/3 ★4/7 = 57% ★1/3 = 33% ★56/98 = 4/7 ★63/189 = 1/3 PAJOILAS, HAZEL ANNE S. | 12 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 ★Rate Nominal ★Codes only represent differences in category, - The frequency of occurrence of a phenomenon in (discrete) class, quality or kind. relation to some specified, useful base number of ★No value (no ranking, just classifying) subjects in a population. ★Rate of occurrence = (p) (a useful base number) Ordinal ★Name categories too but these can be expressed (discrete) as ranks from highest to lowest, etc. - Where p is the proportion of the total group in a category and the useful base number is a multiple of ★Has ranks, has value (highest to lowest) 10. Interval ★Share traits of nominal and ordinal but add a - If in a given large city we find a proportion of (continuous) unit of measure..00000018 deaths by lightning in a year and our base ★Has unit of measure, no true zero value number is 10,000,000, our rate of occurrence is: (.00000018) (10,000,000) or 1.8 lightning deaths per Ratio ★Share traits of interval variables plus a true zero 10,000,000 people. (continuous) point (0=none) ★Rate to Ratio - Let us express a similar idea but this time having the numerator as 1 person (not 1.8 or any number). So we’d like to say 1 is to? ★Formula - Rate/base number = 1/x people - X = (base number / rate) ★Calculation - 1.8/10,000,000 = 1/x people - X = 10,000,000 / 1.8 = 5,555.556 people DATA ★You can improve the level of measurement. ★Variables and Constants - Creating an “index” or “scale” ★Discrete ★Variables and Attributes. - Categorical variables - Attributes are categories. Exclusive and Inclusive. - Categories point out the level of measurement ★Continuous - Scaled variables (infinite) ★Level of measurement and Units of Measure - They may have true zero value - Units of interval-ratio variables only. ★Independent and Dependent Variables ★Coding and counting. Attributes plus. - Intervening and Antecedent Variables (control variable) - [coding] how to transform your data into numbers. ★Hypothesis ROUNDING ERRORS - A prediction about the relationship between two variables, asserting that differences among the ★The difference between the true or perfect score (which we measurements of an independent variable will may never know) and our rounded, observed score. E.g. distance. correspond to differences among the measures of the dependent variable. ★Rounding error depends on what decimal place we choose as our level of precision–our rounding unit. VARIABLES Level of Measurement Procedure for Rounding Round to the nearest point in the line. ★Measurement properties which determine the kind of 1. Specify the rounding unit to its decimal place. mathematical and statistical operations that can be appropriately 2. Observe the number to the right of the rounding units, and used. follow these rules: PAJOILAS, HAZEL ANNE S. | 13 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 a. If it is 0, 1, 2, 3, or 4, round down. ★Cannot be used for nominal level variables b. If it is 6, 7, 8, or 9, round up. c. If it is 5, look at the next decimal place to the right, and FINDING THE MEDIAN if that number is 5 or greater, round up. If there is no 1. Array the cases from low to high (or from high to low) number in the next decimal place, round to an even 2. Locate the middle case number. a. If N is odd: the median is the score of the middle case. ୨ৎSession 7 Discussion: b. If N is even: the median is the average of the scores of the two middle cases. MEASUREMENT OF CENTRAL TENDENCY MEAN ★CHARACTERISTICS ★The average score. - MODE, MEDIAN and MEAN are three different statistics. ★Requires variables measured at the interval-ratio level but is - They report three different kinds of information and often used with ordinal-level variables. will have the same value only in certain situations. ★Cannot be used for nominal-level variables. - They vary in terms of: Level-of-measurement considerations ★The mean or arithmetic average, is by far the most commonly How they define central tendency used measure of central tendency. MODE CHARACTERISTICS OF THE MEAN ★The most common score. ★The mean “balances” out all of the scores in a distribution; all scores “cancel out” around the mean. ★Can be used with variables at all three levels of measurement. ★Most often used with nominal level variables. FINDING THE MODE ★The mean is the point of minimized variation of the scores, “least squares principle” ★Count the number of times each score occurred. ★The score that occurs most often is the mode. a. If the variable is presented in a frequency distribution, ★The mean is affected by all scores; all scores are used in the the mode is the largest category. calculation of the mean b. If the variable is presented in a line chart, the mode is the highest peak. LIMITATIONS OF MODE ★Some distributions have no mode. ★Some distributions have multiple modes. ★The mode of an ordinal or interval-ratio level variable may not be central to the whole distribution. MEDIAN (Md) ★Exact center of distribution of scores ★The score of the middle case ★Can be used with variables measured at the ordinal or interval-ratio levels PAJOILAS, HAZEL ANNE S. | 14 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 The Mean is Affected by All Scores ★Strength - The mean uses all the available information from the variable ★Weaknesses - The mean is affected by every score THE VARIANCE AND THE STANDARD DEVIATION - If there are some very high or low scores (as with skewed distributions), the mean may be mislead ★DEVIATION - distance of any given raw score from its mean. MEAN, MEDIANS, AND SKEW ★Need a measure of variability that takes into account every ★When a distribution has a few very high or low scores, the score. mean will be pulled in the direction of the extreme scores. ★Squaring deviations results in positive calculations. - For a positive skew, the mean will be greater than the ★Square deviations and divide by number of scores. median. - For a negative skew, the mean will be less than the ★The result is the measure known as variance median. ★When an interval-ratio level variable has a pronounced skew, the median may be the more trustworthy measure of central tendency. THE RAW SCORE FOR VARIANCE AND STANDARD MEASURE OF VARIATION DEVIATION ★Range ★There is an easier way using raw scores. - Highest Score - Lowest Score ★Variance THE INTERQUARTILE RANGE ★Inter-quartile range manages effects of extreme outliers ★Standard deviation - IQR = inter-quartile range - Q1 = score at the 1st quartile, 25% below, 75% above - Q3 = score at the 3rd quartile, 75% below, 25% above ★The larger the size of IQR, the greater the variability. ★A type of range measure. SAMPLE AND POPULATION STANDARD DEVIATION ★Considers only the middle 50% of the cases in a distribution. FORMULA ★Avoids some of the problems of the range by focusing on just the middle 50% of scores. ★Sample Standard Deviation Formula ★LIMITATION - Because the Interquartile Range is based on only two scores, it fails to yield any information from all of the other scores. ★Population Standard Deviation Formula STANDARD DEVIATION ★A measure of variation of the scores about the mean. ★The average deviation from the mean is zero. PAJOILAS, HAZEL ANNE S. | 15 SOCIAL RESEARCH METHODS 1 ANNEDOCS | S.Y. 2024-2025 MEASURES OF DISPERSION: SUMMARY ★Higher for more diverse groups (e.g., large samples, populations). ★Decrease as diversity or variety decreases (are lower for more homogeneous groups and smaller samples). ★The lowest value possible for R and s is 0 (no dispersion). PAJOILAS, HAZEL ANNE S. | 16