Courses
What is a Questionnaire, Different types of Questionnaires
what is a questionnaire? A questionnaire is an instrument or research tool consisting of a set of closed-ended or open-ended questions. The purpose is to gather pertinent information from responders for a range of uses. Giving the responder the option to provide a longer response might help them expound on their ideas and provide more insights. The Statistical Society of London created it initially in 1838, and it has been in use ever since. The difference between a Survey and a questionnaire Questions and surveys are now used interchangeably because they have become confused over time. They are two different things. There is a slight but significant difference. The purpose of a questionnaire is to gather information about a person or object. It is not used for trend and pattern identification or statistical analysis. For instance, you may be asked a number of questions about your present physical state when you sign up for a gym membership or during a check-up. By answering these questions, you help us understand your general health, evaluate your risk, and sometimes even identify or diagnose problems. To find trends in a population or to shed light on the larger picture, it is not being used as part of a larger data set. A survey differs slightly. It is used to identify patterns, perform in-depth analysis, and uncover profound insights rather than examining individual questionnaires. Customer feedback surveys, demographic surveys, market research surveys, NPS surveys, etc., are a few examples. The usefulness of these surveys would be significantly reduced if only one person responded. It is simpler to identify trends and reach well-informed conclusions when there are more responders. Businesses can use a QR maker to create QR codes for their surveys or questionnaires, which will further streamline the process of gathering responses. Respondents can easily scan and participate right away by sharing these codes via social media, emails, or posters. Why are they confusing? In the past, the primary users of surveys and questionnaires were researchers and professional marketers. Regarding what they are and when they should be used, they were explicit. In the past, researchers and seasoned marketers were the main users of surveys and questionnaires, having a thorough understanding of their subtleties and knowing exactly when to use them. But as the digital era progresses, the environment has changed significantly. These days, questionnaires and surveys have developed into multipurpose instruments that go beyond conventional limits. These days, they are easily available to companies and individuals looking for answers to confusing consumer behavior, guiding marketing strategies with a burst of creativity that adjusts to the constantly changing market dynamics. Types of Questionnaires? Depending on the type of information you need and its intended use, you can choose between two primary questionnaire types. Questionnaire for exploration (qualitative) They are sometimes referred to as unstructured questionnaires. They are used to gather qualitative data, which is information that isn’t numerical in nature but can be seen and seen. It is applied to characterize and approximate. A qualitative data example would be someone commenting on your writing. They might offer comments on your writing’s tone, clarity, word choice, etc.; this helps you classify it, but you are unable to assign a number to the criticism or feedback. Finding the best writing apps turns into a crucial journey in the field of content creation. These apps give writers a range of tools to manage complexity and embrace stylistic variation, much like perplexity and burstiness define human expression. By experimenting, authors can strike a balance between well-constructed sentences and moments of succinct genius. When you’re just starting out and want to learn more about a topic before coming up with a solution or hypothesis, exploratory questionnaires are perfect. For instance, exploratory questionnaires are perfect if you’re just starting out with product development and don’t know enough about the market. formal, quantitative, standardized questionnaire Another name for them is structured questionnaires. Quantitative data, or information expressed as a count or numerical value, is what these are used to gather. Because the data can be measured, it can be utilized for statistical analysis or mathematical computations. Essentially, it provides an answer to the questions of how much, how many, or how frequently. The best time to use standardized questionnaires is after you have developed a preliminary hypothesis or a product prototype. Before moving forward with product development, you will use it to stress test your hypotheses, designs, use cases, etc. The questions you ask are specific and have a limited scope because of their obvious focus. Types of Questions in a Questionnaire. Not every question type works best in every circumstance. For this reason, it’s critical to first comprehend the kind of questionnaire you’re designing. Selecting the appropriate question types becomes simpler with that knowledge. Open-ended inquiries As the name suggests, the respondent has greater latitude in responding to these questions. The respondent writes as much or as little as they wish rather than being presented with a list of possible answers. For exploratory questionnaires that gather qualitative data, this is perfect. Multi Choice Questions (MCQs) Imagine a questionnaire that serves as your company’s interactive online brochure, drawing in responses while engrossing your target market. Like the consideration that goes into a well-crafted online brochure, keep your overall brand persona and communication in mind as you craft your questions. Dichotomous questions There are only two possible answers to this question. The question is usually yes/no, but it can also be true/false or agree/disagree. Use this when you don’t want to delve too far into the motivations and just need basic validation. Scaled questions In questionnaires, scaled questions are frequently used to assess the intensity of an emotion. Because there are numerous varieties of scaled questions, including the following, this can be utilized in both exploratory and standardized questionnaires: Scale of ratings The Likert scale Semantic differential scale Questions with pictures The last kind of questionnaire question replaces images with text. After answering a question, respondents are presented with a selection of images. Compared to other question types, it typically receives more responses. Likewise, background removal is a technique that can greatly improve the efficiency of visual data collection. Background removal guarantees that respondents’ attention is precisely where it needs to be by separating the subject from any distracting elements. Additionally, using an AI background remover can completely change how questionnaires use images. Clearer and more powerful visual questions are made possible by this tool, which automatically separates the main topic from its background. FAQs Which questions should be avoided in a questionnaire? … Read more
How to determine sample size in research
How to determine sample size in research What should the sample size be, or how big or small should “n” be, is the most pressing question in sampling analysis. The objectives may not be met if the sample size (‘n’) is too small, and we risk significant expenses and resource waste if it is too large. One can generally state that the sample needs to be the ideal size, meaning it shouldn’t be either too big or too small. In theory, the sample size should be sufficiently large to provide a desired width confidence interval; therefore, the sample size must be determined logically before the sample is drawn from the universe. One must focus on the following points in mind: (1) Nature of universe: The universe may be either homogenous or heterogenous in nature. If the items of the universe are homogenous, a small sample can serve the purpose. But if the items are heteogenous, a large sample would be required. Technically, this can be termed the dispersion factor. (ii) Number of classes proposed: If many class groups (groups and sub-groups) are to be formed, a large sample would be required because a small sample might not be able to give a reasonable number of items in each class-group. (ii) Nature of study: If items are to be intensively and continuously studied, the sample should be small. For a general survey the size of the sample should be large, but a small sample is considered appropriate in technical surveys. (V) Accuracy standard and acceptable confidence level: We will need a comparatively larger sample if we want to maintain a high standard of accuracy or precision. The sample size must be increased fourfold in order to double the accuracy for a fixed significance level. (vi) Financial availability: In reality, the sample size is determined by the funds available for the research. This consideration should be made when choosing the sample size because larger samples raise the estimated cost of sampling. (vii) Additional factors to consider: unit type, population size, questionnaire size, and the availability of qualified researchers. Approaches to select sample size: There are two different methods for figuring out the sample size. The first method is used “to specify the precision of estimation desired and then to determine the sample size necessary to insure it” while the second method “uses Bayesian statistics to weigh the cost of additional information against the expected value of the additional information.” The first method is a commonly used method for figuring out ‘n’ because it can provide a mathematical solution. This technique’s drawback is that it fails to compare the expected value of information with the cost of information gathering. Although the second method is theoretically the best, it is rarely employed due to the challenge of determining the information’s value. First of all, it can be said that sampling error always occurs when a sample study is conducted, but it can be managed by choosing a sample that is large enough. The researcher must specify the level of precision he desires for his population parameter estimates. For instance, a researcher may like to estimate the mean of the universe within ±3 of the true mean with 95 percent confidence. In this case we will say that the desired precision is ±3. i.e., if the sample mean is Rs 100, the true value of the mean will be no less than Rs 97 and no more than Rs 103. In other words, all this means that the acceptable error, e, is 3. for different approaches and example: read the upcoming blog do you need the examples? https://tutorarif.com/
Scale of Measurement
levels or Scales of measurement: In research, measurement is the process of assigning numbers to objects or observations, with the level of measurement depending on the rules under which the numbers are assigned. In everyday life, we measure when we use a yardstick to determine the weight, height, or some other feature of a physical object; or when we judge how well we like a song, a painting, or the personalities of our friends. As a result, we measure both physical objects and abstract concepts. For some objects, assigning numbers to their properties is simple, but for others, it can be rather challenging. For example, it is much harder to measure things like social conformity, intelligence, or marital adjustment than it is to measure things like physical weight, biological age, or a person’s financial assets. To put it another way, characteristics like height, weight, and so forth can be measured directly using a standard unit of measurement, but characteristics like drive for success and stress tolerance are more difficult to gauge. When using a yardstick to measure the length of pipe, we can anticipate high accuracy; however, if the idea is abstract and the measurement instruments are not standardized, In technical terms, measurement is the process of applying a correspondence rule to map certain aspects of a domain onto other aspects of a range. In measuring, we create a scale of some kind in the range (range may refer to a set in set theory) and then translate or map the attributes of objects from the domain (domain may refer to another set in set theory) onto this scale. For instance, we could tabulate show attendees by sex if we were doing a study on people who attend a particular performance and wanted to determine the male to female attendance ratio. I assign them a status of 1, 2, 3, or 4, based on whether they are single, Married, widower or divorced. The four levels of measurement (1) Nominal scale: Nominal scale is just a way to label events by giving them numerical symbols. Basketball players are typically identified by their numbers, which serves as an example of this. These numbers are merely convenient labels for the specific class of events and, as such, have no quantitative value. They cannot be regarded as belonging to an ordered scale, nor does their order matter. Keeping track of people, things, and events is made easy with nominal scales. With the numbers involved, there is little that can be done. A group of football players’ numbers, for instance, cannot be effectively averaged to produce a meaningful value. Numbers assigned to one group cannot be usefully compared to numbers assigned to another. When a nominal scale is used, the only arithmetic operation that can be performed is the counting of members in each group. We are therefore limited to using the mode as the central tendency metric. Nominal scales lack a commonly accepted measure of dispersion. The most widely used test of statistical significance is the chi-square test, and the contingency coefficient can be calculated for correlational measures. (2)Ordinal scale: The ordinal scale is the most widely used level of the ordered scale. Events are arranged in order on the ordinal scale, but no effort is made to ensure that the scale’s intervals are equal by any means. Research pertaining to qualitative phenomena commonly uses rank orders, which are ordinal scales. ordinal scale is used to determine a student’s position in his graduating class. When stating scores based on ordinal scales, one must exercise extreme caution. For example, it cannot be claimed that Akram standing is four times better than Ahmed if is ranked 10th in his class and Ahmed is ranked 40. The statement would be completely nonsensical. Items on ordinal scales can only be ranked from highest to lowest. The actual differences between adjacent ranks might not be equal, and ordinal measures lack absolute values. One person’s position on the scale may be higher or lower than another’s, but more accurate comparisons are impossible. Because we are unable to specify how much greater or less, the use of an ordinal scale implies a statement of “greater than” or “less than” (an equality statement is also acceptable). It’s possible that the actual difference between ranks 1 and 2 is less than the difference between ranks 5 and 6. (3) Interval scale: in interval scale the difference between the two consecutive points is the same. Although interval scales can have an arbitrary zero, their main drawback is that they lack a true zero; they are unable to identify what might be referred to as an absolute zero or the unique ongin (ar Responder measure the total absence of a trait or characteristic). An example of an interval scale that illustrates similarities in what can and cannot be done with it is the Fahrenheit scale. One could argue that a temperature increase from 30° to 40° is equivalent to one from 60° to 70°.However, since both temperatures rely on the zero on the scale being set arbitrarily at the freezing point of water, it is impossible to claim that 60° is twice as warm as 30°. The proportion between the two temperatures. (4) Ratio Scale: The actual amounts of variables are represented by a ratio scale. Examples include measurements of physical attributes like height, weight, and distance. In general, any statistical method can be applied to ratio scales, and any manipulation that can be performed on real numbers can also be performed on ratio scale values. This scale can be used for division and multiplication, but not the other scales listed above. Coefficients of variation can also be computed, and geometric and harmonic means can be used as indicators of central tendency. what is NPS Scale? if you know please comment. Face book Page Youtube channel
Basic Terms and concepts in Statistics
Population and Sample: Population Or Statistical Population: a population or a statistical population is a set or a collection of all Objects, and individuals or measurements or recordings, Which may either be finite and or non finite, which has some relevance to some interest or some characteristics, the characteristic may be Heights, weights, ages, mases number etc. the population is said to be of two types (i) Real Population: real population is one whose individuals or recordings are unpredictable with out properly listing them, we need to first record them and know about the individuals For Example: (i) Heights of college students since we will not be able to know about the heights of the students unless we record them one by one (ii) Daily temperature (ii)Hypothetical population: The population whose individuals are predictable even with out actually performing the experiment, or recording the individuals. For Example: (i) A coin tossed since can predict the possible out comes even with out listing them so this is an hypothetical population. Sample: A sample is any subset or any part of a population, the number of individuals and observation in the sample are known as sample size, which is denoted by small “n” Parameter and Statistic: A parameter is any quantity or a value which is calculated from the whole population, a population mean, a population variance are the examples of parameters While a Statistic is any value is calculated from the sample, with out considering the whole population. Observations and variable In Statistics the term Observations refers to some information, which are numerically recorded Or Observations are the recorded values of every element of a characteristic, for example: Physical measurements(Weights and heights) On the other hand variable as name suggests refers to a characteristic which changes it values from individual to individual, place to place and time to time. For example: height, weight, gender are variables and their values taken from different people are observations. Qualitative and Quantitative variables: Qualitative variable: A variable which can not be expressed in numeric form. or a variable whose values or observations can not be numerically recorded. qualitative variable is also know as an Attribute. For example: Blood group, Colors of cars, gender of students etc. Quantitative Variable: Opposite to the qualitative variable we have a quantitative variable, Which refers to a variable which can be numerically expressed. For example: age of the students. income of families. Discrete and Continuous Variables: Discrete variable: discrete variable is a sub type of quantitative variable. It is a variable which can assume or take only discrete (non decimal) or a whole numbers. For example: number of student in a class or a college Income of individuals Continuous variable: A variable which can assume any value fractional or integer in a give range or interval. For example: Age of the students temperature of a place Social Links:
History and Origin of Statistics
Origin Of the word Statistics: The Word “statistics” is said to have been derived either from Latin word “Status” the Italian word “Statista”, or German word “Statistik” these all mean the same as “Political State” or a Government. statistics was originally meant to state or government to know the population size, number of patients in hospitals, number of causalities and injured in any disaster. The roll of statistics was only limited to state holders and rulers and kings who needed to seek information about agricultural land, commerce, military potential, to estimate annual budget, impose tax according to annual expense. in 1602 Shakespeare used the word “Statist” is his drama named Hamlet. The census of the population and recording the trade and rates of the various commodities has been the tradition of the ancient human civilization. In ancient the Roman Empire was one of the first states to collect numerical information called the data. Now a days it is undeniable to accept that the availability of electronics computers is certainly a major factor in the modern development of statistics. And data plays very important role in all walk of life. Computer Technology provided many advantages over calculations by hand or by calculator, stimulated the growth of investigation into new techniques. In the time of Ancient Greece the Philosophers contribute Ideas-no quantitative analyses. In17th Century Jakob Bernoulli, Blais Pascal, Pierre de Fermate, Abraham De Moivre played a very vital role in studying affairs of state, vital statistics of populations probability through games of chance, gambling. In 18th Century the Laplace and Gauss contributed much in regression analysis, correlation and probability distributions importantly in Normal distribution, normal Curve, in connection with the study of astronomy. In 19th Century statistical analysis, regression and corelatioin was first time used in Biology by Quetelet Galton (an astronomer) by studying the genetic variations In 20th century several statistician are active in developing new methods, theories and application of statistics. Pearson Gossett (Student) and Fisher studied natural selection using correlation, and formed first academic department of statistics, Biometrika journal, helped develop the Chi-Square analysis studied process of brewing, alerted the statistics community about problems with small sample sizes, developed Student’s test evolutionary biologists developed ANOVA, stressed the importance of experimental design. Name of some statisticians along with their introduction to different tests and techniques In 20th Century Wilcoxon a biochemist studied pesticides, non-parametric equivalent of two samples test. Kruskalis and Wallis the economists developed the non-parametric equivalent of the ANOVA. Spearman a psychologist developed non-parametric equivalent of the correlation coefficient. Kendall a statistician developed another nonparametric equivalent the correlation coefficient. www.youtube.com/@StatisticsandDataScience123 http://facebook.com/tutorarifsarangzai12 pk.linkedin.com/…/tutor-arif-saranagzai-1ab631323
An Introduction to Statistics
Meaning Of Statistics: Being a very vast subject different people think statistics differently how ever most of the people think of statistics in the following three ways: in General people consider statistics as a subjects which deals with averages, percentages, graphs, charts, and tables. Some other people think that statistics is a subject which consists of some methods, some rules, some techniques, and some formulae, which are used in collecting, and presentation of large amount of numerical information. While other people consider statistics to be a subject which is used to make inference about the population on the basis of sample information. it is interesting to note that inference is made either: (i) On the basis of sample information This is done when it is impossible or it costs much to consider the whole population. then we take a sample which is actually a part of the population and examine it, and then make some inference about the population, this is done in sampling techniques and estimation. (ii) On the basis of past information: this is very effective that we collect past records and data use some technique like regression and others to estimate, and predict about the future outcome. these techniques are useful when there is dependency among the variables. Origin of Statistics: The word “Statistics” is said to be Either Latin root derived form “Status” of Italian word “Statista” these both mean the same as political state or Government. Statistics originally meant information useful to state or government. for example. the size of the population, the number of patient in the hospitals etc. Now a days The word statistics is defined in the following senses: (i ) Statistics in Singular sense: in singular senses statistics means a subject which is a body of some methods, techniques, rules and formulae. which are useful in collection, presentation, analysis, and interpretation of Numerical facts or data. (ii) Statistics in plural sense: In Plural Sense the word “statistics” refers to “numerical information” or numerical facts. for example: Statistics of prices (data or information related to prices) Statistics of road accidents. (data or information related to road accidents) statistics of educational institutions: Statistics in plural sense can also be called “Statistical Data” https://tutorarif.com/the-difference-between-filers-and-non-filers/ http://linkedin.com/in/tutor-arif-saranagzai-1ab631323 are you interested in knowing history of statistics? keep visiting the webite