PR (2nd Quarter) PDF

PR (2nd QUARTER) -jillsc sobrang quote HOW DO WE COMPUTE FOR SAMPLE? Stratified Sampling - small subgroups (strata) - randomly selected from each of these...

PR (2nd QUARTER) -jillsc sobrang quote HOW DO WE COMPUTE FOR SAMPLE? Stratified Sampling - small subgroups (strata) - randomly selected from each of these strata. Systematic Sampling - Has sequence - Chosen at regular intervals Sampling ALTERNATIVE WAY = USING - Divided into clusters RAOSOFT - Clusters are randomly selected ABOUT RAOSOFT (WHEN TO USE?) Multi-stage Sampling - Combination of one or more methods 1% Margin of Error: - Population is divided into multiple clusters and then these clusters are Requires a very large sample size because it further divided and grouped into demands extremely precise results. various sub groups (strata) based on similarity. One or more clusters can be 5% Margin of Error: randomly selected from each stratum. Most common for surveys, balancing This process continues until the precision and practicality. cluster can't be divided anymore. 10% Margin of Error: Convenience Sampling - Availability Requires fewer respondents but is less precise - Easy to deliver results 99% if life ng tao Purposive Sampling - Intention or purpose of the study 95% if standard confidence level Quota Sampling - Proportion of characteristics/ traits in 90% pag malaki yung sample size the sample should be the same as the population. SAMPLING TECHNIQUES Snowball/Referral Sampling PROBABILITY - equal chances - Completely unknown and rare NON PROBABILITY - biased - Help from first element THE KINDS AND TYPES OF INSTRUMENTS SAMPLING TECHNIQUES - tools used to gather data for a particular research topic. Simple Random Sampling - Equal chances - No prior information PR (2nd QUARTER) -jillsc sobrang quote Ways to develop an instrument for quantitative research Content validity: Is the test fully representative of what it aims to measure? - Adopting an instrument - Modify an existing instrument A mathematics teacher develops an - Create your own instrument end-of-semester algebra test for her class. The test should cover every form of algebra TYPES OF QUESTIONS that was taught in the class. YES OR NO Construct validity: Does the test measure the RECOGNITION TYPE concept that it's intended to measure? COMPLETION TYPE CODING TYPE You develop a questionnaire to diagnose SUBJECTIVE TYPE - What can you say depression, so you need to know: does the about the teachers who are deeply committed questionnaire really measure the construct of to their work? depression? Or is it actually measuring the COMBINATION TYPE - Is a combination respondent's mood, self-esteem, or some other of two or more type of questions. construct? LIKERT SCALE - respondents are asked how much they agree or disagree with a set of Criterion validity: Do the results correspond statements. to a different test of the same thing? SEMANTIC DIFFERENTIAL SCALE - - A university professor creates a new test to require respond to rate their attitude by measure applicants' English writing ability. To selecting a position on a bipolar adjective scale assess how well the test really does measure students' writing ability, she finds an existing test that is considered a valid measurement of VALIDITY AND REALIBILITY English writing ability and compares the results when the same group of students take VALIDITY both tests. If the outcomes are very similar, the -refers to the ability of an instrument to new test has a high criterion validity. measure -what it intends to measure. Two Types of Criterion Validity RELIABILTY -refers to the consistency of results and a Concurrent Validity: An instrument has this reliable instrument yields the same results for validity when it is able to predict results individuals who take the test more than once similar to those of a test already validated in the past. Types of Instrument Validity Predictive Validity: When an instrument Face validity: Does the content of the test produces results similar to those of another appear to be suitable to its aims? instrument that will be employed in the future? You created a survey to measure the regularity of people's dietary habits. You review the survey items, which ask questions about every meal of the day and snacks eaten in between for every day of the week. PR (2nd QUARTER) -jillsc sobrang quote 3 Types of Reliability AFTER Test-retest reliability 1. Immediately encode and archive your data. [also known as Stability] measures the 2. Safeguard the confidentiality of your data. consistency of results when you repeat the 3. Later, examine and analyze your raw data same test on the same sample at a different using appropriate statistical tools. point in time. QUANTITATIVE DATA ANALYSIS Parallel forms reliability measures the correlation between two QUANTITATIVE RESEARCH equivalent versions of a test. You use it when The data is usually gathered using structured you have two different assessment tools or sets research instruments. of questions designed to measure the same The research study can usually be replicated thing. or repeated, given its high reliability. Data are in the form of numbers and Inter-rater reliability statistics, often arranged in tables, charts, (also called interobserver reliability) figures, other non-textual forms. measures the degree of agreement between different people observing or assessing the STATISTICS same thing. You use it when data is collected collecting and analyzing numerical data by rescarchers assigning ratings, scores to one performing some arithmetic procedures or more variable. DESCRIPTIVE STATISTICS DATA COLLECTION PROCEDURE Describes a certain aspect of data set by calculating Mean (average), Median BEFORE (midpoint), and Mode (common) 1. Develop your data collection instruments It tells about the placement or position of one and materials. data item in relation to the other data, the 2. Seek permission from the authorities and extent of the distribution or spreading out of heads of the institutions/communities where data, and whether they are correlations or you will conduct your study. regressions between or among variables. 3. Select and screen the population using This kind of statistics does not tell anything appropriate sampling techniques. about the population. 4. Obtain informed consent from the respondents. Mean: numerical average of a set of values. 5. Pilot-test the instruments to determine Median: midpoint of a set of numerical potential problems that may occur when they values. are administered. Mode: most common value among a set of values. Percentage: used to express how a value or DURING group of respondents within the data relates to a larger group of respondents. 1. Provide instructions to the respondents and Frequency: the number of times a value is explain how the data will be collected. found. 2. Administer the instruments and implement Range: the highest and lowest value in a set of the intervention or treatment (if applicable). values PR (2nd QUARTER) -jillsc sobrang quote INFERENTIAL STATISTICS T-TEST Focuses conclusions, generalizations, below 30 sample size predictions, interpretations, hypotheses. parametric statistical technique that tests the There are a lot of hypotheses testing in this difference between two means method of statistics that require you to perform T-test for two dependent samples - same complex and advanced mathematical groups are highly related to each other or in operations. other words SAME subjects (pre-test and post-test) Bivariate Analysis - analysis of two T-test for two independent samples - it variables (independent and dependent tests the difference between data sets from two variables) different groups such as in the case of control - T-test, z-test, pearson's and treatment groups. Multivariate Analysis - analysis of multiple relations between multiple variables STEPS - ANOVA UNPAIRED T- TEST 1. Data PARAMETRIC TEST 2. Descriptive stats - T-test 3. Get the mean, standard deviation, - Z-test variance - ANOVA 4. Divide higher variance to lower NON-PARAMETRIC TEST variance (0-4 = equal; 5-8 = unequal) - Chi-square 5. "T-test two samples assuming __ - Mann-Whitney variances" - Kruskal-Wallis 6. Hypothesized mean difference (subtract 2 means) Parametric test - distribution is known and 7. Get p value based on a fixed set of parameters Nonparametric test - distribution of PAIRED T-TEST population is not known, and parameters are 1. Data not fixed 2. Descriptive stats Mean - refers to the average score of a given 3. Highlight mean, standard deviation, set of values (sum of all/ divided by the total variance number of values). 4. "T-test paired two sample for means" Variance - refers to how spread out the 5. Hypothesized mean difference values are across the data set you are studying. (subtract 2 means) it helps you to find how close or not close the 6. Get p value data to the mean Standard Deviation - square root of the Z- TEST variance. Used for more than 30 sample size Alpha level (significance level) - refers to the probability value that must be reached STEPS TO COMPUTE USING Z- TEST: before claiming that findings obtained are 1. Data statistically significant 2. Descriptive stats P-value - calculated probability that is 3. Highlight mean, standard deviation, compared to the alpha level variance PR (2nd QUARTER) -jillsc sobrang quote 4. "Z-test two samples for means" CO-EFFICIENT SCALE 5. Hypothesized mean difference (subtract 2 means) 6. Variance 1 and 2 7. Get p value ANOVA statistical tool used for testing differences SCATTER PLOT among the means of two or more groups of Use to present the results of samples. Pearson's r visually. it considers both the variation within and Best graph for presenting correlation between the sample groups between two variables. One-way ANOVA - tests differences among groups concerning one variable. Two-way ANOVA - used for determining the relationship between TWO independent nominal variables and ONE dependent interval or continuous variable. it finds out whether only one or both independent variables cause changes in the dependent variable. REMINDERS IN COMPUTING ANOVA Compute ANOVA: Remember this: p-value < alpha level Decision Making: Reject the Null Hypothesis Failed to Reject the Null Hypothesis F crit › f = reject; F crit < = do not reject PEARSON'S R is parametric statistical method used for determining whether there is a linear relationship between variables. Positive correlation - when the numerical value of one variable increases or decreases, the other variable increases or decreases as well Negative correlation - when the numerical value of one variable increases, the other variable decreases, and vice-versa. No correlation - when the two variable have no relationship with each other.