Statistics involves collecting, organizing, and analyzing numerical data. There are two main types: descriptive statistics, which involves organizing and summarizing sample data, and inferential statistics, which uses sample data to draw conclusions about populations. Key concepts in statistics include populations, samples, parameters, statistics, variables, and different types of data and sampling methods. Measures of central tendency like the mean, median, and mode describe the center of data sets, while measures of variability describe how spread out data is. Reliability and validity are also important statistical concepts used to evaluate measurement methods.
Statistics involves collecting, organizing, and analyzing numerical data. There are two main types: descriptive statistics, which involves organizing and summarizing sample data, and inferential statistics, which uses sample data to draw conclusions about populations. Key concepts in statistics include populations, samples, parameters, statistics, variables, and different types of data and sampling methods. Measures of central tendency like the mean, median, and mode describe the center of data sets, while measures of variability describe how spread out data is. Reliability and validity are also important statistical concepts used to evaluate measurement methods.
Statistics involves collecting, organizing, and analyzing numerical data. There are two main types: descriptive statistics, which involves organizing and summarizing sample data, and inferential statistics, which uses sample data to draw conclusions about populations. Key concepts in statistics include populations, samples, parameters, statistics, variables, and different types of data and sampling methods. Measures of central tendency like the mean, median, and mode describe the center of data sets, while measures of variability describe how spread out data is. Reliability and validity are also important statistical concepts used to evaluate measurement methods.
STATISTICS - A branch of mathematics that Measurement.
examines and investigates ways to process and Experimental Classification analyze the date gathered. A researcher may classify - The study of how to collect, organize, variables according to the analyze, and interpret numerical function they serve in the information from data. experiment. Descriptive Statistics Mathematical Classification - Involves methods of organizing, Variables may also be picturing, and summarizing classified in terms of the information from samples mathematical values. Inferential Statistics Independent Variables - Involves methods of using Variables controlled by the information from a sample to experimenter, and expected to draw conclusions have an effect on the behavior of Population the subjects. Consists of all members of Dependent Variable the group. Is some measure of the behavior Sample of subjects and expected to be A portion, or part, of the influenced by the independent population of interest variable. selected for analysis. Continuous Variable Parameter A variable which can assume any Numerical index describing of an infinite number of values. a characteristic of a Discrete Variable population. A variable which consist of either a Statistics finite number of values or Numerical index describing countable number of values. a characteristic of a Nominal sample. Used to differentiate Primary Data classes or categories Data that comes from the original for purely classification. source, and are intended to Ordinal answer specific questions. This applies to data Secondary Data that can be arranged Data that are taken from in order. previously recorded data. Interval Constant classify order A characteristic that does not vary. and differentiate Variable between classes or A characteristic that can take categories in terms of different values. degrees of differences. Qualitative Variable There is no natural zero Describes an individual by starting point category or group. Ratio Quantitative Variable Applies to data that Has a value or numerical can be arranged in order. There is a true zero. - Snowball Sampling Random Sampling is a technique in which one or more Type of sampling in which members of a population are located all members have equal and chance of being selected. used to lead the researchers to other - Simple Random Sampling members of the population A process whose members had an - Voluntary Sampling equal A technique when samples are chance of being selected from the composed of population via random numbers. respondents who are self-select in the - Systematic Sampling study/survey A process of selecting a kth element in the population until the desired number of subjects or respondents is attained - Stratified Sampling Subdivide the population into at least SUMMARY MEASURES two CENTRAL TENDENCY different subgroups that share the same a summary measure that attempts to describe a characteristics, then draw a sample whole set of data with a single value that from represents the middle or centre of its each subgroup (or stratum) distribution. - Cluster Sampling MEAN Divide the population into sections (or It is the average of our data set and it is clusters); randomly select some of those easily affected by outliers. clusters; Outlier Non-Random Sampling is a data point that differs significantly from A sampling procedure other where samples selected in observations. a deliberate manner with MEDIAN little or no attention to It is the middle value and arranged in randomization order. It is the opposite of Mean since it won’t - Convenience Sampling be affected by outliers because it is in the A process of selecting a group of middle and it can’t be changed. individuals MODE who (conveniently) are available for Occurs most frequently. It is the most study commonly occurring value in a - Purposive Sampling distribution. A process of selecting based from MIDRANGE judgment It is the middle point of a range of to select a sample which the researcher numbers. Average of the lowest and believed, based on prior information, highest scores. will POINT MEASURES provide the data they need. give us a way to see where a certain data point - Quota Sampling or value falls in a sample or distribution. A Is applied when an investigator survey measure can tell us whether a value is about the collects information from an assigned average, or whether it’s a value falls is an number, or quota of individuals outlier. MEASURES OF VARIABILITY PERCENTILE RANK are statistics that describe the - percentage of scores in its frequency amount of difference and spread in a distribution that are equal to or lower data set. than it. RANGE RELIABILITY The difference between the largest and the refers to how consistently a method measures smallest measures. something. If the same result can be INTERQUARTILE RANGE (IQR) consistently achieved by using the same Half the distance between quartile methods under Points measures the spread of your data's the same circumstances, the measurement is middle half. considered reliable. VARIANCE Test-retest a measure of dispersion a measure of reliability obtained by that takes into account the spread of administering the same test twice over a period all data points in a data set. of time to a group of individuals. STANDARD DEVIATION Internal consistency A measure of how dispersed the data is which is the consistency of people’s responses in relation to the mean. across the items on a multiple-item measure. Inter-rater reliability is the extent to which different observers are consistent in their judgments. Parallel Forms Reliability Measures the correlation between two equivalent versions of a test. You use it when Z- SCORES you have two different assessment tools or sets (also known as a standard scores) of questions designed to measure the same - Helps to understand where a score lies thing. in relation to other scores in the Split-half reliability distribution is a measure of consistency whereby a set of - Helps to understand where a score lies items that make up a measure is split in two in relation to other scores in the during the data distribution Standard Score VALIDITY allows us to calculate the probability of a score is the extent to which the data or results occur within our normal distribution and of a research method represent enables us to compare two scores that are from the intended variable. different normal CONSTRUCT VALIDITY distributions. Evaluates whether a measurement tool really DEVIATION IQ SCORES represents the thing we are interested in a standard score with a mean of 100 and a measuring. It’s central to standard establishing the overall validity of a method. deviation of 15. CONTENT VALIDITY T- SCORES Assesses whether a test is representative of all are standard scores with a mean of 50 and a aspects of the construct. standard deviation of 10. FACE VALIDTY SCALED SCORES considers how suitable the content of a test are standard scores with a mean of 10 and a seems to be on the surface. It’s similar to standard deviation of 3. content validity, but face validity is a more Lambda informal and subjective assessment. • test used to measure relationship between CRITERION VALIDITY two nominal variables evaluates how well a test can predict a concrete Gamma outcome, or how well the results of your test • test used to measure relationship between approximate the results of another test. two ordinal variables Eta NORMAL DISTRIBUTION • test used to measure relationship between form a bell-shaped curved that is symmetric interval and nominal variables about a vertical line through the mean of the Linear Regression data. • test used to predict the dependent variable SKEWNESS from the given (one) independent a measure of the asymmetry of a variable. distribution. A distribution is asymmetrical Multiple Linear Regression when • test used to predict the dependent variable its left and right side are not mirror images. from several independent variables. POSITIVE SKEWED t-test for one sample (One sample t-test) the mean of the data is greater than the median • used to compare population mean to sample (a large number of data-pushed on the right- mean hand side). t-test for dependent sample NEGATIVE SKEWED • test used to compare means from the same The mean of the data is less than the median (a groups large Example number of data-pushed on the left-hand side). t-test for independent sample KURTOSIS • test used to compare means from two a measure of whether the different (independent) groups data are heavy-tailed or light-tailed One-way ANOVA relative to a normal distribution. • test used to compared three or more means. Leptokurtic or heavy-tailed distribution Chi square (x2) (kurtosis more than normal distribution). • test between two nominal variables. Mesokurtic (kurtosis) Same as the normal distribution). Non-Parametric Test Platykurtic or short-tailed Distribution • if one of the assumptions for parametric test (kurtosis less than normal distribution). was violated Parametric Test • if the data are normally distributed (Shapiro- Binomial Test Wilk Test/Kolmogorov-Smirnov Test) • A non parametric test used to compare two • more than 30 samples nominal data (dichotomous) • probability sampling technique was used to Spearman rho select the sample • A non parametric test used to measure relationship (correlation) between interval Pearson’s r variables. • test used to measure relationship (correlation) Wilcoxon Signed Ranked Test between interval variables • A non parametric test use to compare two Point Biserial dependent/related groups • test used to measure relationship (correlation) Mann Whitney U Test between interval and nominal • A non parametric test use to compare two (dichotomous) variables. independent/different groups Kruskal Wallis H Test NULL HYPOTHESIS • A non parametric test used to compared three -the hypothesis that we hope to reject or do not or more means. reject. ALTERNATIVE HYPOTHESIS Summary (Counterparts) -the hypothesis that will be accepted. 1. chi-square – Binomial Test TWO TAILED 2. Pearson’s r - Spearman rho -Used for non directional hypothesis. 3. t-test for dependent sample – Wilcoxon ONE TAILED signed rank test -used for directional hypothesis. 4. t-test for independent sample – Mann TYPE 1 ERROR Whitney U test -rejected when in fact it is true (FALSE POSITIVE) 5. One-way ANOVA – Kruskal Wallis H Test TYPE II ERROR -not rejected when in fact it is false(FALSE STANDARD ERROR NEGATIVE) -used to estimate the standard deviation of the sampling distribution DIFFERENCE OF PROPORTIONS -used to check if the data is reliable and stable. CONFIDENCE INTERVAL -is the range of values that you expect your estimate to fall between a certain percentage of the time you ran the experiment. CONFIDENT LEVEL -is the percentage of times you expect to reproduce an estimate between the upper and lower bound of confidence level. POINT ESTIMATE -will be whatever statistical estimate you are making. CRITICAL VALUE -tells you how many standard deviation away from the mean you need to go in order to reach the desired confidence level to you confidence interval. LEVEL OF SIGNIFICANCE -it is defined if the null hypothesis is assumed to be accepted or rejected. INFERENTIAL STATISTICS -the process of making inferences or generalizations on population based on the results of the study on the samples. HYPOTHESIS -assumption made about the probability distribution of the population.