Data analysis techniques
Online training this is also our fresher session for our beloved teachers to enhance their skills and technology every saturday we will conduct webinar sessions for teachers about a use advantages and relevance of different blended learning software applications on webinars hearing it morning sessions you can watch us in our depth at edtech unit facebook page educational technology unit youtube channel and dab at philippines high school learners and parents who are watching our live stream of our italia online tutorial regional research coordinator at region 8 mom jenna lindaya and to our regional learning resource coordinator mr joy biha good afternoon mom jennilyn and sir joy and also i would like to greet dinpo and among our research valerie charlie izamay maria gloria julia stacy julian lorenzo christine wilfred renan okay malaya june at samahaniako class as we learn very important topics for inquiries investigations and immersion and today's exciting session will lead us to understand our topic on quarter four week 3 which is on data analysis techniques specifically at the end of this tutorial session you are going to apply the appropriate quantitative data analysis technique to be used in the given problem and for the information of everyone we are basing our tutorial session today with the presentations of sir chris anthony r sabino a senior high school the division of ramblan a region 4b mimarupa and sir james pedrera from alan alan national high school division of late region 8. to begin i would like you class to perform this preliminary activity entitled classifying techniques what we're going to do in here is that you are going to classify each word showing the different data analysis techniques commonly used in research you are going to classify which of them relates to quantitative or qualitative research then write one if it relates to quantitative research and qual if it relates to qualitative research again right one if it is quantitative and qual if qualitative for now i want you to observe first and study the words flashed on your screen that is to know which words relate to quantitative or qualitative research and later when i'm going to mention each word or each data analysis technique you have five seconds to write your answers in our live streams comment section and also please do not forget clash to indicate the item that you are answering all right are you ready let us begin number one frequency counts is it one or qual again number one frequency counts is this word or analysis technique relates to quantitative research or qualitative research julian hazel maria gloria stacy all answered quan that is correct quantitative number two mean mean is it quantitative or qualitative and you are correct number two is for quantitative beatrice april you are correct how about number three narrative analysis is it quan or qual all right julian answered qual andrea also answered qual that is correct narrative analysis is a qualitative analysis technique how about number four variance is it quan or qual jane onesie answered quan that is correct very good okay number five discourse analysis is it quantitative or qualitative beatriz answered qual for number five that is correct this course analysis is a qualitative analysis technique for number six content analysis is it quan or qual raf radores answered qual for discourse analysis how about for content analysis number six jean kenneth izamay maria gloria answered qual you are correct content analysis is a qualitative analysis technique number seven percentage is it quan or qual jin kenneth answered quan for number seven hazel rose also answered quan as well as maria gloria jane quincy micah and rafa doris answered quan that is correct percentage is a quantitative data analysis technique how about number eight thematic is it quan or qual jean kenneth answered qual that is correct thematic is a qualitative analysis technique number nine t-test for dependent samples is it for quan or qual number nine john v gabas answered kwan okay as well as julian well fred ah wait number eight well fred answered qual j.
R juan izamay andrea julia hazel rose giancarlo beatrice all answered 1 for number 9 and you are all correct t test for dependent samples is used for quantitative research last item number 10 pearson r or pearson product moment correlation is it used for quantitative data analysis technique or qualitative number 10 okay a lot answered quan starting from hazel rose chris andrea giancarlo julia jane wency charlie rafadores and a lot more so pearson r is of course a quantitative adita analysis technique so congratulations class for answering it all correctly how did you find activity class was it easy or difficult how are you able to classify the words or what helped you in classifying the words and class when do we appropriately use this data analysis technique in c quantitative qualitative data analysis technique all right take note class that each type of research whether qualitative or quantitative has its own distinguishing data analysis techniques as student researchers it is necessary that we know the meaning and purpose of these data analysis techniques and class this is what this is what we're going to discuss in our tutorial session today let us start our discussion with the different data analysis techniques used in quantitative research because for our next tutorial session we will discuss the different data analysis techniques used in qualitative research all right quantitative data analysis technique class is divided into two we have descriptive and inferential statistics but first let me talk about descriptive statistics descriptive statistics are used in describing a set of data if you may recall in our previous tutorial session that in order for us to answer our descriptive research questions like questions asking the profile of respondents in terms of age and gender or to answer the question like the question what is the senior high school students level of dengue knowledge all right to answer or analyze these descriptive type of research questions of course we will use descriptive statistics descriptive statistics includes frequency count percentage measures of central tendency the mean weighted mean median and mode as well as the measures of variability range variance and standard deviation let us start first with frequency counts frequency counts measure or frequency counts measure the number of times that an event or observation occurs example for frequency count is the counting of number let's say number of male and number of female respondents in your study or to know what are the responding sources of information about dengue okay since the answer to this question can be obtained through counting the students responses as their source of dengue knowledge another descriptive analysis technique is through the use of percentage percentage is actually a part no it's a part of a whole distribution like if you want to know in your study how many percent are males or how many percent are females all right another is the mean or the computational average or the average of the given data set this is the most commonly used descriptive statistics in educational research like in order for you to answer the question what is the senior high school students level of dengue knowledge you need to obtain or measure the mean or the average score of the respondents based on the respondent score or the student score in a test about dengue right then we also have what we call weighted mean okay weighted mean can be used to know the perception of the respondents based on their answers in a likert in a like likert type or a likert scale or a likert type questionnaire you may say in a likert type questionnaire or in a likert scale um students okay uh their scores can be multiplied with the weights no in every uh descriptor or in every scale for example we multiply times one times three times five and then you get the weighted average we also have the median which is also called as the positional average it is the middlemost value when data are arranged in ascending or descending order then the mode also referred to as the nominal average which which is also known as the most common value or the value that appears most often in a data set like for example to answer the question what tv channel is the most commonly watched by senior high school students of course that can be answered through the use of mode and take note class that mean median and mode are measures of central tendency while the most common measures of variability or measure of dispersion is the range standard deviation and variance the range is just the difference between the largest and smallest values in a set of data or the highest and lowest value in a data set next or another measure of dispersion is the variance okay the variance is actually this the squared value of your standard deviation so the square root of your variance is just your standard deviation wherein the standard deviation is the measure of dispersion of a set of data from its mean or from the average in here we can measure the homogeneity and the heterogeneity of the data or we will be able to know whether the scores in the data set is homogenous or not no through the use of the measure of variability or the measure of this person all right now to compute the script the descriptive statistics using the microsoft office data analysis tool pack specifically in the microsoft excel okay what you're going to do is you're going to okay you're going to follow these steps all right so unlike in your statistics and probability class that you're going to really compute no manually the the measures or the values to get for let's say to get the value for your mean for your standard deviation or for your variance but in doing research okay so we we have we have actually a lot of tools no our software's that we can use no but uh not nothing is for analysis all right so when you're going to do um or use the ms data analysis toolpak so first is you're going to click data then look for descriptive statistics and click ok next is you're going to enter your input and output range or you can just simply highlight the data for your input and output then check summary statistics and then click ok all right so last is you're going to click ok so where you're able to follow class so for you to practice please try computing the descriptive statistics of your own study using this microsoft office data analysis tool pack all right now what if you are asked class these questions or what if you will have these research questions in your study like is there a significant difference between the first and second semester general average grade of stem learners or is there a significant relationship between the preferred learning modality and the academic performance of senior high school learners in the new normal okay what do you think is the essential statistics to be used in this type of questions significant difference significant relationships okay these questions can be answered inferential statistics or inferential statistical techniques through hypothesis testing all right so what is inferential statistics in inferential statistics data are analyzed from a sample to make inferences in the larger collection of the population the purpose is to answer or test your research hypothesis thus hypothesis testing is a procedure for making rational decisions about the reality of observed facts okay hypothesis uh when used in plural form is called hypotheses no two or more hypotheses okay hypothesis or hypotheses is a proposed explanation for a phenomenon it has two types we have two types of hypotheses null and alternative hypothesis first step in hypothesis testing is you are going to state the null and the alternative hypothesis a null hypothesis denoted by h sub o is a denial of existence of significant difference effect or relationship or which states that there is no significant difference effect or relationship between or among variables actually in simpler terms it is the hypothesis that the researcher tries to disprove no in the study on the other hand alternative alternative hypothesis denoted by h sub a or h sub 1 is an affirmation of existence of significant difference effect or relationship or which states that there is a significant difference effect or relationship among variables take note class that failure to reject the null hypothesis or h sub o does not mean that the null hypothesis is true or we are going to accept the null hypothesis it only means that we do not have sufficient evidence to support the alternative hypothesis or the h sub 1 all right and here are the statistical treatments that we can use to test for difference tests for relationship and tests for association as you can see in this table that we have two types of tests and non-parametric tests we will use non-parametric tests when your data potentially does not follow a normal distribution or if your data is not or at least not interval in nature like for example if your data is ordinal or you have a rank data and when there is no randomness of your samples all right so then if that is the case you will use non-parametric tests explore all right nothing explores possibility before tayo gumamit of using parametric tests so we will use t tests okay in order to know the difference of two independent groups t tests if given that the sample size is less than or equal to 30 z tests if the sample size is greater than 30.
Then we will use anova or analysis of variance right if you have three or more independent groups pearson r to test the relationship of two variables or pearson product moment correlation okay we can use pearson r to test the relationship of two variables with an interval or ratio data right then we also have a simple and multiple regression to test for association for the purpose of this tutorial session let us say assumptions is a test for normality and based from the results it says that the data is normally distributed homogeneity and your data are mutually exclusive or there is independence and given that you are dealing with interval or ratio data with large data sets so when all of these assumptions were met so we can proceed to using parametric tests right another step in hypothesis testing is to determine the appropriate statistical technique to be used t test of independent samples in which it determines whether the means or the averages of two independent groups differ significantly for example when you are going to compare a data or data is compared or the data compared are coming from two groups let us say male and female group or if you want to compare the scores of students using an intervention and the scores of students without using an intervention so the appropriate test to be used is t tests for independent samples okay meaning it can only be used in two unrelated groups on the other hand t tests of dependent samples determines whether the means of the two dependent groups differ significantly this is also referred to as t tests for paired samples in here the data are coming from one group for example you want to compare the score before and after an intervention or if you want to compare the pre-tests and post test scores of one group let us say the experimental group so the dependent t test or paired sample t test can be used to test either a change or a difference in the means between two related groups all right to determine if there is a significant relationship between variables or to test the correlation between variables we can use pearson's r or pearson product moment correlation which is used to measure the strength or magnitude and the direction of the linear relationship of two quantitative variables take note two quantitative variables so that can be interval or ratio data right the spearman rank correlation or spearman's row evaluates the monotonic relationship between two ordinal variables or variables that can be rank okay remember class that the last step in hypothesis testing is to write the conclusion but we cannot decide whether to reject or do not reject the null hypothesis or the h sub o without using these methods first is the use of the p-value method all right if the p-value okay take note if the p-value is less than or equal to the level of significance or alpha wherein in educational research the commonly used level of significance is 5 or 0.
05 so if the p value is less than or equal to 0.05 then we reject the null hypothesis it means that the difference or the relationship is significant however if the p-value is greater than the level of significance or if the p-value is greater than 0.05 then we fail to reject the null hypothesis it means that the difference or the relationship is not significant all right the second approach that we can use in our decision making is the use of the critical value approach if the computed value take note if the computed value whether t computed or z computed is greater than the tabular or critical value then we reject the null hypothesis so opposite point p value approach okay so take note huh kanina if less lesser than or equal to the the if the p value is lesser than or equal to the level of significance we reject the null hypothesis if using the critical value approach okay given that okay if the computed decomputed or z computed capability is greater than the tabular or critical value then we reject the null hypothesis it means that the difference or relationship is significant however if the computed value is lesser than or equal to the tabular or critical value then we fail to reject the null hypothesis it means that the difference or relationship is not significant right now in order for us to interpret our relationship or the relationship of variables we can use this interpretation table for correlation using bess and khan 2006 and cohen 1992 so for best and can 2006 no if your computed r or computed correlation coefficient ranges from zero to plus minus 0.
20 that is negligible plus minus point 20 to positive negative point 40 that is low plus minus point forty two plus minus point sixty that is moderate plus minus point sixty up to plus minus zero point eighty that's substantial plus minus point eighty up to plus minus 1.0 meaning high to very high relationship using cohen's 1992 okay table for correlation we can have negative 0.3 to positive 0.3 there is a weak relationship okay negative 0.5 to negative 0.3 or 0.3 to 0.5 there is a moderate relationship negative 0.9 to negative 0.5 or 0.5 to 0.9 there is a strong relationship between the variables negative 1.0 to negative 0 up 0.09 or that is point negative 0.9 or 0.9 to 1.0 there is a very strong relationship now using mendenhall beaver r and beaver b 2013 okay to test if the really uh if the result no for correlation is significant or the relationship is significant or not so we can use this table for our interpretation now if the p-value is less than 0.01 then our interpretation is highly significant p value is greater than or equal to 0.01 or it ranges from 0.01 to 0.05 so our relationship is significant now if the p-value is greater than 0.05 then the relationship is not significant all right now let us try using this tables in our example let's have a problem the given problem says a dietetic student wanted to look at the relationship between calcium intake and knowledge about calcium in sports science students or those students in the special program for sports all right now we have here our data here are the knowledge score or the score of students obtained no in a test about calcium calcium intake or the knowledge about calcium given that we have 20 respondents out of 50 items so here are their scores and we also have the uh the calcium intake no the calcium intake of the same set of respondents those 20 respondents in the same order okay from 450 1050 then we also have 1085 all right now we will not be manually computing right the uh value for the correlation all right now we have the microsoft excel for us to do that okay so what we're going to do what we're going to focus is how we're going to interpret if we already have the the result if you already have the data how are we going to interpret that no because if we know how to compute for the correlation we know how to compute for the tests for uh let's say test for means no if well if we don't know how to interpret or how to analyze these results okay because given this data of the computations my interpret given the result of your computation all right now let's say that our research question for that problem is is there a significant relationship between calcium intake and knowledge about calcium in sports science students so we can create an important so given that research question we can have our null hypothesis to be there is no significant relationship between calcium intake and knowledge about calcium in sports science students or equivalent to saying the correlation coefficient r is just equal to zero now we have the alternative hypothesis the opposite of the null hypothesis so we state there is a significant correlation or relationship between calcium intake and knowledge about calcium in sports science students equivalent to saying r or the correlation coefficient is not equal to zero right so therefore what is the correlation coefficient r so if we're going to use the microsoft excel we're going to type or key in our data and then look for a pearson r okay so we will have this result okay knowledge score the relation this is our correlation matrix all right so knowledge to knowledge score of course that's a perfect relationship one calcium intake to calcium intake of course that is one so it's a no young relationship knowledge score to calcium intake so we can see it here that is 0.
88 okay so we will just use the uh the two decimal places so approximately equal to 0.88 now what is the nature and strength of our relationship given that the correlation coefficient r is equal to 0.88 so therefore we will use our okay wait if we go back to best anchor okay best anchor 2016 dito and cohen's 1992 so 0.88 is actually that is high to very high and then using cohen that is there is a strong relationship all right so can you take note no important young table nothing no young table for uh interpretation of the correlation or the relationship all right what is the p-value okay so so result nothing's uh microsoft excel the p-value is less than 0.001 so using the p-value is the relationship significant or not so we go back to mendenhall's at al interpretation table okay we we will try to look if given the p-value now the p-value dow is less than 0.001 so i think that is zero point zero zero zero zero zero something no raised to the nth power okay now so if the p value is less than okay 0.00 therefore that is highly significant all right so using mendenhall at al 2013 okay therefore we can say that the relationship is significant okay in here less than 0.01 how much more if that is 0.001 all right now what is your decision against your null hypothesis so given that the relationship is significant okay using the uh coherence 1992 interpretation as well as using the mendenhall's interpretation table therefore we can say that the null hypothesis is stating that there is no significant correlation or relationship between calcium intake and knowledge about calcium in sports science students or students in the special program for sports equivalent to saying r is equal to 0 is rejected okay so we reject the null hypothesis all right so here is the epa style of reporting the correlation or the relationship okay given r okay r of 20 meaning you have 20 samples is equal to 0.88 okay your correlation coefficient is equal to 0.
88 and your p is less than or p-value or probability value is less than 0.001 so we can write knowledge about calcium and calcium intake are strongly and significantly correlated magnitude at magnitude meaning strong relationship tapos that the relationship is significant okay so r of 20 is equal to 0.88 or p less than 0.001 the p-value associated with the test statistic is highly significant meaning there is a sufficient evidence against the null hypothesis hence it is rejected it implies that in the population high calcium intake is associated with high knowledge about kashum all right so nineteen di and puba young interpretation all right now aside from determining if the relationship or the correlation is significant or not or if the relationship is strong no meaning we determine the magnitude of the relationship we can also determine the direction of the relationship whether there is a positive correlation or a negative correlation all right so using paren your microsoft excel so as you can see in the figure okay meron is the knowledge score okay of 20 respondents out of 50 items and then your y and y nathan and then click no this one this uh scatter plot so we can have that a figure all right so okay let's have a closer look of that figure okay so we have this okay calcium intake and knowledge about calcium right so in the x-axis we have the knowledge score okay out of 50 items okay and then the calcium intake of 20 students right in the y-axis so if you go back to our this one this is the cutter scatter plots and correlation examples okay so if the scatter plot will look like this meaning there is a perfect positive correlation if it looks like this okay not perfectly but closer young dots from uh left to right it is increasing all right so there is a high or highly positive correlation okay uh if it is uh decreasing no from left to right it is decreasing therefore there is a high negative correlation right now for low positive correlation as you can see hindi masado uh young manga that's nothing so there is a low positive correlation but still meron ring direction it is going up no from left to right the dots are going up so there is still paren a positive correlation perro if the dots from left to right is decreasing so there there is a low negative correlation now for no correlation as you can see direction okay so that will be the example of um a scatter plot of uh the relationship of the variables wherein there is no correlation okay so therefore given our scatter plot this one so therefore we can say that there is a high positive correlation as you can see straight line from left to right no increasing from left to right okay so there's a high positive correlation so what does it mean when there is a high positive correlation meaning there's a high positive correlation between the calcium intake and the knowledge about calcium among students in the special program for sports so if there is a high positive correlation that implies that if their knowledge about calcium increases right then they will more likely increase also their calcium intake and the other way around all right 19d han pumbaa okay now now that you have gained a clear understanding on your or in our topic today let us apply what you've learned by answering the following activity you are going to write the appropriate data analysis technique or techniques okay to be used for the following problems please write your answers in our live streams comment section all right are you ready number one teachers perception on the implementation of senior high school using four point scale what um data analysis technique quantitative data analysis technique can be used in this research okay again number one teacher's perception on the implementation of senior high school using four-point scale or likert scale jane wensitadora answered descriptive statistics that is correct janewensey specifically that is we can use weighted mean for that all right or specifically for um to answer the perceptions of the teachers using a four-point likert scale so we can use weighted mean all right now for number two the relationship between the first and second semester grades of stem learners again number two the relationship between the first and second semester grades of stem learners what quantitative data analysis technique can we use for item number two so since we are looking for the relationship between the variables all right correct stacy very good pearson r micah hindu answered inferential statistics that is correct also specifically we will use pearson r or pearson product moment correlation then for last item number three the difference between pre-tests and poses results in mathematics okay of course since we want to know the difference between the pre-test and poses results in mathematics of course we will be using inferential statistics so specifically what uh data analysis technique under inferential statistical for number three again pre-test and post-test results meaning same group difference okay same group okay then we look into the difference between their pre-test and their post-test scores independent samples a t test for dependent samples or t tests for paired samples alindo on yon gagamitinaten for item number three for independent samples right t test for dependent samples hazel rose answered t test for dependent samples sir that is correct or dependent t tests answered by maria gloria that is correct also or we can say we can use t tests for paired samples okay t tests for dependent samples or dependent sample t tests or paired sample t tests all right bucket paired sample because jung scores nothing says okay then we try to look into their pre-test the difference between the pre-test and post-test scores all right 19-digit then we go to number four item number four okay you are going to write the letter of your answer in the comment section all right a researcher want or wants to find out if time spent on facebook and friendliness are correlated correlation analysis revealed a correlation coefficient of 0.
88 and a p-value of 0.052 given this therefore we can say that the relationship found is blank a highly significant b not significant c significant the very significant mighty answer b sir not significant maria gloria judea also answered letter b and you are correct okay using mendenhall's table for correlation interpretation we have okay since the p-value is 0.052 so it falls less than okay zero less than zero point i mean greater than 0.05 okay 0.052 is greater than 0.05 therefore we say that the relationship or the correlation is not significant very good last item right you are going to interpret the following correlation coefficient using cohen's interpretation 1992 for the given r value or the given correlation coefficient the correlation coefficient is 0.70 okay so what will be the magnitude or the strength of the correlation given or using coherence 19 1992 interpretation table so you write the word no in the uh comment section is there a weak relationship moderate strong or very strong migi answered strong stacy also as is strong very good so given the correlation coefficient of 0.70 and using coherence interpretation table since 0.70 falls in the interval 0.5 to 0.9 therefore we say that there is a strong relationship or there is a strong correlation between the given variables all right nineteenth class all right now for your assignment since this is three eyes you are going to apply the appropriate data analysis technique that you are going to use given your research questions okay the by meronataion research questions that you have formulated from the preview sessions now following also the guidelines presented in the tutorial class kindly post okay in your fb timeline and use the hashtag hashtag e2li iii analyzing data again the hashtag is hashtag etholi iii analyzing data weekly live tutorial sessions you.
May follow me so investigations and immersion are compiled in our wakelet account at ianpo i'm online tutorial salingogito see you po in our next tutorial session same time same day this has been your ito light shooter sir june nanaki the research of today shall speak the innovations of tomorrow thank you and happy researching is is on foreign is is.
R juan izamay andrea julia hazel rose giancarlo beatrice all answered 1 for number 9 and you are all correct t test for dependent samples is used for quantitative research last item number 10 pearson r or pearson product moment correlation is it used for quantitative data analysis technique or qualitative number 10 okay a lot answered quan starting from hazel rose chris andrea giancarlo julia jane wency charlie rafadores and a lot more so pearson r is of course a quantitative adita analysis technique so congratulations class for answering it all correctly how did you find activity class was it easy or difficult how are you able to classify the words or what helped you in classifying the words and class when do we appropriately use this data analysis technique in c quantitative qualitative data analysis technique all right take note class that each type of research whether qualitative or quantitative has its own distinguishing data analysis techniques as student researchers it is necessary that we know the meaning and purpose of these data analysis techniques and class this is what this is what we're going to discuss in our tutorial session today let us start our discussion with the different data analysis techniques used in quantitative research because for our next tutorial session we will discuss the different data analysis techniques used in qualitative research all right quantitative data analysis technique class is divided into two we have descriptive and inferential statistics but first let me talk about descriptive statistics descriptive statistics are used in describing a set of data if you may recall in our previous tutorial session that in order for us to answer our descriptive research questions like questions asking the profile of respondents in terms of age and gender or to answer the question like the question what is the senior high school students level of dengue knowledge all right to answer or analyze these descriptive type of research questions of course we will use descriptive statistics descriptive statistics includes frequency count percentage measures of central tendency the mean weighted mean median and mode as well as the measures of variability range variance and standard deviation let us start first with frequency counts frequency counts measure or frequency counts measure the number of times that an event or observation occurs example for frequency count is the counting of number let's say number of male and number of female respondents in your study or to know what are the responding sources of information about dengue okay since the answer to this question can be obtained through counting the students responses as their source of dengue knowledge another descriptive analysis technique is through the use of percentage percentage is actually a part no it's a part of a whole distribution like if you want to know in your study how many percent are males or how many percent are females all right another is the mean or the computational average or the average of the given data set this is the most commonly used descriptive statistics in educational research like in order for you to answer the question what is the senior high school students level of dengue knowledge you need to obtain or measure the mean or the average score of the respondents based on the respondent score or the student score in a test about dengue right then we also have what we call weighted mean okay weighted mean can be used to know the perception of the respondents based on their answers in a likert in a like likert type or a likert scale or a likert type questionnaire you may say in a likert type questionnaire or in a likert scale um students okay uh their scores can be multiplied with the weights no in every uh descriptor or in every scale for example we multiply times one times three times five and then you get the weighted average we also have the median which is also called as the positional average it is the middlemost value when data are arranged in ascending or descending order then the mode also referred to as the nominal average which which is also known as the most common value or the value that appears most often in a data set like for example to answer the question what tv channel is the most commonly watched by senior high school students of course that can be answered through the use of mode and take note class that mean median and mode are measures of central tendency while the most common measures of variability or measure of dispersion is the range standard deviation and variance the range is just the difference between the largest and smallest values in a set of data or the highest and lowest value in a data set next or another measure of dispersion is the variance okay the variance is actually this the squared value of your standard deviation so the square root of your variance is just your standard deviation wherein the standard deviation is the measure of dispersion of a set of data from its mean or from the average in here we can measure the homogeneity and the heterogeneity of the data or we will be able to know whether the scores in the data set is homogenous or not no through the use of the measure of variability or the measure of this person all right now to compute the script the descriptive statistics using the microsoft office data analysis tool pack specifically in the microsoft excel okay what you're going to do is you're going to okay you're going to follow these steps all right so unlike in your statistics and probability class that you're going to really compute no manually the the measures or the values to get for let's say to get the value for your mean for your standard deviation or for your variance but in doing research okay so we we have we have actually a lot of tools no our software's that we can use no but uh not nothing is for analysis all right so when you're going to do um or use the ms data analysis toolpak so first is you're going to click data then look for descriptive statistics and click ok next is you're going to enter your input and output range or you can just simply highlight the data for your input and output then check summary statistics and then click ok all right so last is you're going to click ok so where you're able to follow class so for you to practice please try computing the descriptive statistics of your own study using this microsoft office data analysis tool pack all right now what if you are asked class these questions or what if you will have these research questions in your study like is there a significant difference between the first and second semester general average grade of stem learners or is there a significant relationship between the preferred learning modality and the academic performance of senior high school learners in the new normal okay what do you think is the essential statistics to be used in this type of questions significant difference significant relationships okay these questions can be answered inferential statistics or inferential statistical techniques through hypothesis testing all right so what is inferential statistics in inferential statistics data are analyzed from a sample to make inferences in the larger collection of the population the purpose is to answer or test your research hypothesis thus hypothesis testing is a procedure for making rational decisions about the reality of observed facts okay hypothesis uh when used in plural form is called hypotheses no two or more hypotheses okay hypothesis or hypotheses is a proposed explanation for a phenomenon it has two types we have two types of hypotheses null and alternative hypothesis first step in hypothesis testing is you are going to state the null and the alternative hypothesis a null hypothesis denoted by h sub o is a denial of existence of significant difference effect or relationship or which states that there is no significant difference effect or relationship between or among variables actually in simpler terms it is the hypothesis that the researcher tries to disprove no in the study on the other hand alternative alternative hypothesis denoted by h sub a or h sub 1 is an affirmation of existence of significant difference effect or relationship or which states that there is a significant difference effect or relationship among variables take note class that failure to reject the null hypothesis or h sub o does not mean that the null hypothesis is true or we are going to accept the null hypothesis it only means that we do not have sufficient evidence to support the alternative hypothesis or the h sub 1 all right and here are the statistical treatments that we can use to test for difference tests for relationship and tests for association as you can see in this table that we have two types of tests and non-parametric tests we will use non-parametric tests when your data potentially does not follow a normal distribution or if your data is not or at least not interval in nature like for example if your data is ordinal or you have a rank data and when there is no randomness of your samples all right so then if that is the case you will use non-parametric tests explore all right nothing explores possibility before tayo gumamit of using parametric tests so we will use t tests okay in order to know the difference of two independent groups t tests if given that the sample size is less than or equal to 30 z tests if the sample size is greater than 30.
Then we will use anova or analysis of variance right if you have three or more independent groups pearson r to test the relationship of two variables or pearson product moment correlation okay we can use pearson r to test the relationship of two variables with an interval or ratio data right then we also have a simple and multiple regression to test for association for the purpose of this tutorial session let us say assumptions is a test for normality and based from the results it says that the data is normally distributed homogeneity and your data are mutually exclusive or there is independence and given that you are dealing with interval or ratio data with large data sets so when all of these assumptions were met so we can proceed to using parametric tests right another step in hypothesis testing is to determine the appropriate statistical technique to be used t test of independent samples in which it determines whether the means or the averages of two independent groups differ significantly for example when you are going to compare a data or data is compared or the data compared are coming from two groups let us say male and female group or if you want to compare the scores of students using an intervention and the scores of students without using an intervention so the appropriate test to be used is t tests for independent samples okay meaning it can only be used in two unrelated groups on the other hand t tests of dependent samples determines whether the means of the two dependent groups differ significantly this is also referred to as t tests for paired samples in here the data are coming from one group for example you want to compare the score before and after an intervention or if you want to compare the pre-tests and post test scores of one group let us say the experimental group so the dependent t test or paired sample t test can be used to test either a change or a difference in the means between two related groups all right to determine if there is a significant relationship between variables or to test the correlation between variables we can use pearson's r or pearson product moment correlation which is used to measure the strength or magnitude and the direction of the linear relationship of two quantitative variables take note two quantitative variables so that can be interval or ratio data right the spearman rank correlation or spearman's row evaluates the monotonic relationship between two ordinal variables or variables that can be rank okay remember class that the last step in hypothesis testing is to write the conclusion but we cannot decide whether to reject or do not reject the null hypothesis or the h sub o without using these methods first is the use of the p-value method all right if the p-value okay take note if the p-value is less than or equal to the level of significance or alpha wherein in educational research the commonly used level of significance is 5 or 0.
05 so if the p value is less than or equal to 0.05 then we reject the null hypothesis it means that the difference or the relationship is significant however if the p-value is greater than the level of significance or if the p-value is greater than 0.05 then we fail to reject the null hypothesis it means that the difference or the relationship is not significant all right the second approach that we can use in our decision making is the use of the critical value approach if the computed value take note if the computed value whether t computed or z computed is greater than the tabular or critical value then we reject the null hypothesis so opposite point p value approach okay so take note huh kanina if less lesser than or equal to the the if the p value is lesser than or equal to the level of significance we reject the null hypothesis if using the critical value approach okay given that okay if the computed decomputed or z computed capability is greater than the tabular or critical value then we reject the null hypothesis it means that the difference or relationship is significant however if the computed value is lesser than or equal to the tabular or critical value then we fail to reject the null hypothesis it means that the difference or relationship is not significant right now in order for us to interpret our relationship or the relationship of variables we can use this interpretation table for correlation using bess and khan 2006 and cohen 1992 so for best and can 2006 no if your computed r or computed correlation coefficient ranges from zero to plus minus 0.
20 that is negligible plus minus point 20 to positive negative point 40 that is low plus minus point forty two plus minus point sixty that is moderate plus minus point sixty up to plus minus zero point eighty that's substantial plus minus point eighty up to plus minus 1.0 meaning high to very high relationship using cohen's 1992 okay table for correlation we can have negative 0.3 to positive 0.3 there is a weak relationship okay negative 0.5 to negative 0.3 or 0.3 to 0.5 there is a moderate relationship negative 0.9 to negative 0.5 or 0.5 to 0.9 there is a strong relationship between the variables negative 1.0 to negative 0 up 0.09 or that is point negative 0.9 or 0.9 to 1.0 there is a very strong relationship now using mendenhall beaver r and beaver b 2013 okay to test if the really uh if the result no for correlation is significant or the relationship is significant or not so we can use this table for our interpretation now if the p-value is less than 0.01 then our interpretation is highly significant p value is greater than or equal to 0.01 or it ranges from 0.01 to 0.05 so our relationship is significant now if the p-value is greater than 0.05 then the relationship is not significant all right now let us try using this tables in our example let's have a problem the given problem says a dietetic student wanted to look at the relationship between calcium intake and knowledge about calcium in sports science students or those students in the special program for sports all right now we have here our data here are the knowledge score or the score of students obtained no in a test about calcium calcium intake or the knowledge about calcium given that we have 20 respondents out of 50 items so here are their scores and we also have the uh the calcium intake no the calcium intake of the same set of respondents those 20 respondents in the same order okay from 450 1050 then we also have 1085 all right now we will not be manually computing right the uh value for the correlation all right now we have the microsoft excel for us to do that okay so what we're going to do what we're going to focus is how we're going to interpret if we already have the the result if you already have the data how are we going to interpret that no because if we know how to compute for the correlation we know how to compute for the tests for uh let's say test for means no if well if we don't know how to interpret or how to analyze these results okay because given this data of the computations my interpret given the result of your computation all right now let's say that our research question for that problem is is there a significant relationship between calcium intake and knowledge about calcium in sports science students so we can create an important so given that research question we can have our null hypothesis to be there is no significant relationship between calcium intake and knowledge about calcium in sports science students or equivalent to saying the correlation coefficient r is just equal to zero now we have the alternative hypothesis the opposite of the null hypothesis so we state there is a significant correlation or relationship between calcium intake and knowledge about calcium in sports science students equivalent to saying r or the correlation coefficient is not equal to zero right so therefore what is the correlation coefficient r so if we're going to use the microsoft excel we're going to type or key in our data and then look for a pearson r okay so we will have this result okay knowledge score the relation this is our correlation matrix all right so knowledge to knowledge score of course that's a perfect relationship one calcium intake to calcium intake of course that is one so it's a no young relationship knowledge score to calcium intake so we can see it here that is 0.
88 okay so we will just use the uh the two decimal places so approximately equal to 0.88 now what is the nature and strength of our relationship given that the correlation coefficient r is equal to 0.88 so therefore we will use our okay wait if we go back to best anchor okay best anchor 2016 dito and cohen's 1992 so 0.88 is actually that is high to very high and then using cohen that is there is a strong relationship all right so can you take note no important young table nothing no young table for uh interpretation of the correlation or the relationship all right what is the p-value okay so so result nothing's uh microsoft excel the p-value is less than 0.001 so using the p-value is the relationship significant or not so we go back to mendenhall's at al interpretation table okay we we will try to look if given the p-value now the p-value dow is less than 0.001 so i think that is zero point zero zero zero zero zero something no raised to the nth power okay now so if the p value is less than okay 0.00 therefore that is highly significant all right so using mendenhall at al 2013 okay therefore we can say that the relationship is significant okay in here less than 0.01 how much more if that is 0.001 all right now what is your decision against your null hypothesis so given that the relationship is significant okay using the uh coherence 1992 interpretation as well as using the mendenhall's interpretation table therefore we can say that the null hypothesis is stating that there is no significant correlation or relationship between calcium intake and knowledge about calcium in sports science students or students in the special program for sports equivalent to saying r is equal to 0 is rejected okay so we reject the null hypothesis all right so here is the epa style of reporting the correlation or the relationship okay given r okay r of 20 meaning you have 20 samples is equal to 0.88 okay your correlation coefficient is equal to 0.
88 and your p is less than or p-value or probability value is less than 0.001 so we can write knowledge about calcium and calcium intake are strongly and significantly correlated magnitude at magnitude meaning strong relationship tapos that the relationship is significant okay so r of 20 is equal to 0.88 or p less than 0.001 the p-value associated with the test statistic is highly significant meaning there is a sufficient evidence against the null hypothesis hence it is rejected it implies that in the population high calcium intake is associated with high knowledge about kashum all right so nineteen di and puba young interpretation all right now aside from determining if the relationship or the correlation is significant or not or if the relationship is strong no meaning we determine the magnitude of the relationship we can also determine the direction of the relationship whether there is a positive correlation or a negative correlation all right so using paren your microsoft excel so as you can see in the figure okay meron is the knowledge score okay of 20 respondents out of 50 items and then your y and y nathan and then click no this one this uh scatter plot so we can have that a figure all right so okay let's have a closer look of that figure okay so we have this okay calcium intake and knowledge about calcium right so in the x-axis we have the knowledge score okay out of 50 items okay and then the calcium intake of 20 students right in the y-axis so if you go back to our this one this is the cutter scatter plots and correlation examples okay so if the scatter plot will look like this meaning there is a perfect positive correlation if it looks like this okay not perfectly but closer young dots from uh left to right it is increasing all right so there is a high or highly positive correlation okay uh if it is uh decreasing no from left to right it is decreasing therefore there is a high negative correlation right now for low positive correlation as you can see hindi masado uh young manga that's nothing so there is a low positive correlation but still meron ring direction it is going up no from left to right the dots are going up so there is still paren a positive correlation perro if the dots from left to right is decreasing so there there is a low negative correlation now for no correlation as you can see direction okay so that will be the example of um a scatter plot of uh the relationship of the variables wherein there is no correlation okay so therefore given our scatter plot this one so therefore we can say that there is a high positive correlation as you can see straight line from left to right no increasing from left to right okay so there's a high positive correlation so what does it mean when there is a high positive correlation meaning there's a high positive correlation between the calcium intake and the knowledge about calcium among students in the special program for sports so if there is a high positive correlation that implies that if their knowledge about calcium increases right then they will more likely increase also their calcium intake and the other way around all right 19d han pumbaa okay now now that you have gained a clear understanding on your or in our topic today let us apply what you've learned by answering the following activity you are going to write the appropriate data analysis technique or techniques okay to be used for the following problems please write your answers in our live streams comment section all right are you ready number one teachers perception on the implementation of senior high school using four point scale what um data analysis technique quantitative data analysis technique can be used in this research okay again number one teacher's perception on the implementation of senior high school using four-point scale or likert scale jane wensitadora answered descriptive statistics that is correct janewensey specifically that is we can use weighted mean for that all right or specifically for um to answer the perceptions of the teachers using a four-point likert scale so we can use weighted mean all right now for number two the relationship between the first and second semester grades of stem learners again number two the relationship between the first and second semester grades of stem learners what quantitative data analysis technique can we use for item number two so since we are looking for the relationship between the variables all right correct stacy very good pearson r micah hindu answered inferential statistics that is correct also specifically we will use pearson r or pearson product moment correlation then for last item number three the difference between pre-tests and poses results in mathematics okay of course since we want to know the difference between the pre-test and poses results in mathematics of course we will be using inferential statistics so specifically what uh data analysis technique under inferential statistical for number three again pre-test and post-test results meaning same group difference okay same group okay then we look into the difference between their pre-test and their post-test scores independent samples a t test for dependent samples or t tests for paired samples alindo on yon gagamitinaten for item number three for independent samples right t test for dependent samples hazel rose answered t test for dependent samples sir that is correct or dependent t tests answered by maria gloria that is correct also or we can say we can use t tests for paired samples okay t tests for dependent samples or dependent sample t tests or paired sample t tests all right bucket paired sample because jung scores nothing says okay then we try to look into their pre-test the difference between the pre-test and post-test scores all right 19-digit then we go to number four item number four okay you are going to write the letter of your answer in the comment section all right a researcher want or wants to find out if time spent on facebook and friendliness are correlated correlation analysis revealed a correlation coefficient of 0.
88 and a p-value of 0.052 given this therefore we can say that the relationship found is blank a highly significant b not significant c significant the very significant mighty answer b sir not significant maria gloria judea also answered letter b and you are correct okay using mendenhall's table for correlation interpretation we have okay since the p-value is 0.052 so it falls less than okay zero less than zero point i mean greater than 0.05 okay 0.052 is greater than 0.05 therefore we say that the relationship or the correlation is not significant very good last item right you are going to interpret the following correlation coefficient using cohen's interpretation 1992 for the given r value or the given correlation coefficient the correlation coefficient is 0.70 okay so what will be the magnitude or the strength of the correlation given or using coherence 19 1992 interpretation table so you write the word no in the uh comment section is there a weak relationship moderate strong or very strong migi answered strong stacy also as is strong very good so given the correlation coefficient of 0.70 and using coherence interpretation table since 0.70 falls in the interval 0.5 to 0.9 therefore we say that there is a strong relationship or there is a strong correlation between the given variables all right nineteenth class all right now for your assignment since this is three eyes you are going to apply the appropriate data analysis technique that you are going to use given your research questions okay the by meronataion research questions that you have formulated from the preview sessions now following also the guidelines presented in the tutorial class kindly post okay in your fb timeline and use the hashtag hashtag e2li iii analyzing data again the hashtag is hashtag etholi iii analyzing data weekly live tutorial sessions you.
May follow me so investigations and immersion are compiled in our wakelet account at ianpo i'm online tutorial salingogito see you po in our next tutorial session same time same day this has been your ito light shooter sir june nanaki the research of today shall speak the innovations of tomorrow thank you and happy researching is is on foreign is is.