Master of Business Administration - Semester 3 MB 0050: “Research Methodology” (4 credits) (Book ID: B1206) ASSIGNMENT- Set 2 Marks 60 Note: Each

Question carries 10 marks
Q 1. What is the significance of research in social and business sciences? Ans: According to a famous Hudson Maxim, “All progress is born of inquiry. Doubt is often better than overconfidence, for it leads to inquiry, and inquiry leads to invention”. It brings out the significance of research, increased amounts of which makes progress possible. Research encourages scientific and inductive thinking, besides promoting the development of logical habits of thinking and organization. The role of research in applied economics in the context of an economy or business is greatly increasing in modern times. The increasingly complex nature of government and business has raised the use of research in solving operational problems. Research assumes significant role in formulation of economic policy, for both the government and business. It provides the basis for almost all government policies of an economic system. Government budget formulation, for example, depends particularly on the analysis of needs and desires of the people, and the availability of revenues, which requires research. Research helps to formulate alternative policies, in addition to examining the consequences of these alternatives. Thus, research also facilitates the decision making of policy-makers, although in itself it is not a part of research. In the process, research also helps in the proper allocation of a country’s scare resources. Research is also necessary for collecting information on the social and economic structure of an economy to understand the process of change occurring in the country. Collection of statistical information though not a routine task, involves various research problems. Therefore, large staff of research technicians or experts is engaged by the government these days to undertake this work. Thus, research as a tool of government economic policy formulation involves three distinct stages of operation which are as follows:

Market research is refers to the investigation of the structure and development of a market for the formulation of efficient policies relating to purchases. and analytical techniques to find solution to business problems such as cost minimization or profit maximization. production and sales. meaning that the supposed answer is leaning towards an outcome. It gives intellectual satisfaction of knowing things for the sake of knowledge. What is the meaning of hypothesis? Discuss the types of hypothesis. It is also the starting point for discussing the important parts of a study. Research is equally important to social scientist for analyzing social relationships and seekingexplanations to various social problems. mathematical. All these researches are very useful for business and industry. research in social sciences is concerned with both knowledge for its own sake. i. and The prognosis. It is also known as an educated guess. and that it is also based on some . the prediction of future developments Research also assumes a significant role in solving various operational and planning problems associated with business and industry. The key characteristic of any hypothesis is the fact that it is testable. A good hypothesis should state a prediction. market research. operations research. A hypothesis can be a single proposition or be made up of several propositions which will trigger a set of scientific experiments to prove right or wrong. or the optimization problems.e. A hypothesis is vital because it helps shape the research structure intended to solve a problem or find an answer. It also possesses practical utility for the social scientist to gain knowledge so as to be able to do something better or in a more efficient manner. In several ways.. It is important for research in any field because it limits a researcher’s view of problems. which are responsible for business decision making. Q 2. and knowledge for what it can contribute to solve practical problems. Ans: A hypothesis is an answer derived from limited evidence in order to start another investigation to explain an event. it is concerned with the analyzing the motivations underlying consumer behavior. and allows them to examine certain aspects of a problem.• • • Investigation of economic structure through continual compilation of facts Diagnoses of events that are taking place and the analysis of the forces underlying them. phenomena. More specifically. Operational research relates to the application of logical. as there is no assumption of truth involved. This. Motivational research helps to determine why people behave in the manner they do with respect to market characteristics. or mathematical model. and motivational research are vital and their results assist in taking business decisions.

Probability and nonprobability sampling A probability sampling scheme is one in which every unit in the population has a chance (greater than zero) of being selected in the sample. Other factors that should be taken into consideration when writing a hypothesis are its simplicity. As a remedy. For example. Types of hypothesesa.states that relation or difference between variables exists. it is possible to identify and measure every single item in the population and to include any one of them in our sample. Next. we seek a sampling frame which has the property that we can identify every single element and include any in our sample. Nondirectional . there is no way to identify which people will actually vote at a forthcoming election (in advance of the election). expands. Deductive is derived from theory and provides evidence that supports. possible sampling frames include an electoral register and a telephone directory. The most straightforward type of frame is a list of elements of the population (preferably the entire population) with appropriate contact information.states that there is no significant relation or difference between variables. Null . e. make an educated guess on the direction of the difference. There is no way to identify all rats in the set of all rats. d. the more actionable the results for the succeeding experiments.previous observations. such as the sentencing of a batch of material from production (acceptance sampling by lots). Where voting is not compulsory.states the expected direction of the relation or difference. in the more general case this is not possible. and conservatism. a researcher must first identify what the problem is. and this . Q 3. However. c. in an opinion poll. b. When formulating one. These imprecise populations are not amenable to sampling in any of the ways below and to which we could apply statistical theory. or contradicts the theory. Explain the sampling process Ans: Sampling frame In the most straightforward case. Directional . The more specific a hypothesis. Inductive is a generalization based on specific observations. fruitfulness. scope.

so we simply add their income to our estimate of the total.probability can be accurately determined. Systematic Sampling. Involves random selection at some point. To reflect this. generated from a uniform distribution between 0 and 1. and interview the first person to answer the door. this is a nonprobability sample. by weighting sampled units according to their probability of selection. Hence. which forms the criteria for selection. Probability sampling includes: Simple Random Sampling. (For example. because the selection of elements is nonrandom. These conditions give rise to exclusion bias. nonprobability sampling does not allow the estimation of sampling errors. Example: We visit every household in a given street. placing limits on how much information a sample can provide about the population. We then interview the selected person and find their income. Every element has a known nonzero probability of being sampled and 2. because some people are more likely to answer the door (e. not everybody has the same probability of selection. or where the probability of selection can't be accurately determined.g. an unemployed person who spends most of their time at home is more likely to answer than an employed housemate who might be . Stratified Sampling. (The person who is selected from that household can be loosely viewed as also representing the person who isn't selected. what makes it a probability sample is the fact that each person's probability is known. In any household with more than one occupant. and randomly select one adult from each household. Nonprobability sampling is any sampling method where some elements of the population have no chance of selection (these are sometimes referred to as 'out of coverage'/'undercovered'). we would count the selected person's income twice towards the total. The combination of these traits makes it possible to produce unbiased estimates of population totals. when we come to such a household. this is known as an 'equal probability of selection' (EPS) design. People living on their own are certain to be selected. These various ways of probability sampling have two things in common: 1. We visit each household in that street. and Cluster or Multistage Sampling. we can allocate each person a random number. Example: We want to estimate the total income of adults living in a given street. It involves the selection of elements based on assumptions regarding the population of interest. making it difficult to extrapolate from the sample to the population. Such designs are also referred to as 'self-weighting' because all sampled units are given the same weight. identify all adults living there.) In the above example. Probability Proportional to Size Sampling. But a person living in a household of two adults has only a one-in-two chance of selection. and select the person with the highest number in each household). When every element in the population does have the same probability of selection. Information about the relationship between sample and population is limited.

a simple random sample of ten people from a given country will on average produce five men and five women. all such subsets of the frame are given an equal probability. nonresponse effects may turn any probability design into a nonprobability design if the characteristics of nonresponse are not well understood. discussed below. SRS can be vulnerable to sampling error because the randomness of the selection may result in a sample that doesn't reflect the makeup of the population. In addition. For example. individually or in combination. In particular. This minimises bias and simplifies analysis of results. For instance. Each element of the frame thus has an equal probability of selection: the frame is not subdivided or partitioned. SRS cannot accommodate the needs of researchers in this situation because it . However. and so on). the variance between individual results within the sample is a good indicator of variance in the overall population. and the need to measure accuracy Whether detailed analysis of the sample is expected Cost/operational concerns Simple random sampling In a simple random sample ('SRS') of a given size. Factors commonly influencing the choice between these designs include: • • • • • Nature and quality of the frame Availability of auxiliary information about units on the frame Accuracy requirements. Furthermore. investigators are interested in research questions specific to subgroups of the population. In some cases. quota sampling and purposive sampling.at work when the interviewer calls) and it's not practical to calculate these probabilities. a variety of sampling methods can be employed. researchers might be interested in examining whether cognitive ability as a predictor of job performance is equally applicable across racial groups. which makes it relatively easy to estimate the accuracy of results. but any given trial is likely to overrepresent one sex and underrepresent the other. any given pair of elements has the same chance of selection as any other such pair (and similarly for triples. attempt to overcome this problem by using information about the population to choose a more representative sample. since nonresponsive effectively modifies each element's probability of being sampled. SRS may also be cumbersome and tedious when sampling from an unusually large target population. Nonprobability sampling methods include accidental sampling. Sampling methods Within any of the types of frame identified above. Systematic and stratified techniques.

A simple random selection of addresses from this street could easily end up with too many from the high end and too few from the low end (or vice versa). Example: Suppose we wish to sample people from a long street that starts in a poor district (house #1) and ends in an expensive district (house #1000). Stratified sampling. cheap side. leading to an unrepresentative sample. making the scheme less accurate than simple random sampling. but is instead randomly chosen from within the first to the kth element in the list. which is discussed below. and the even-numbered houses are all on the south (cheap) side. If periodicity is present and the period is a multiple or factor of the interval used. Selecting (e. the sample is especially likely to be unrepresentative of the overall population. Systematic sampling involves a random start and then proceeds with the selection of every kth element from then onwards. Under the sampling scheme given above. It is easy to implement and the stratification induced can make it efficient. Systematic sampling Systematic sampling relies on arranging the target population according to some ordering scheme and then selecting elements at regular intervals through that ordered list. (Note that if we always start at house #1 and end at #991.does not provide subsamples of the population. if the variable by which the list is ordered is correlated with the variable of interest. 'Every 10th' sampling is especially useful for efficient sampling from databases. In this case. or they will all be from the evennumbered. Simple random sampling is always an EPS design (equal probability of selection). It is important that the starting point is not automatically the first in the list. systematic sampling is especially vulnerable to periodicities in the list. this bias is eliminated. k=(population size/sample size). by randomly selecting the start between #1 and #10.) However.g. either the houses sampled will all be from the odd-numbered. As long as the starting point is randomized. addresses this weakness of SRS. but not all EPS designs are simple random sampling. also referred to as 'sampling with a skip of 10'). systematic sampling is a type of probability sampling. A simple example would be to select every 10th name from the telephone directory (an 'every 10th' sample. representing all of these districts. the sample is slightly biased towards the low end.) every 10th street number along the street ensures that the sample is spread evenly along the length of the street. it is impossible' to get a representative sample. expensive side. Example: Consider a street where the odd-numbered houses are all on the north (expensive) side of the road. .

out of which individual elements can be randomly selected.Another drawback of systematic sampling is that even in scenarios where it is more accurate than SRS. see discussion of PPS samples below. First. the frame can be organized by these categories into separate "strata. one in ten). .but because this method never selects two neighbouring houses. There are several potential benefits to stratified sampling... Systematic sampling can also be adapted to a non-EPS approach. Second. different sampling approaches can be applied to different strata.994} has a one-in-ten probability of selection. instead of availability of the samples). for an example.g. but the set {4. Finally. it is sometimes the case that data are more readily available for individual.. pre-existing strata within a population than for the overall population. the sample will not give us any information on that variation. provided that each stratum is proportional to the group's size in the population. dividing the population into distinct. It is not 'simple random sampling' because different subsets of the same size have different selection probabilities . independent strata can enable researchers to draw inferences about specific subgroups that may be lost in a more generalized random sample.34.} has zero probability of selection. Even if a stratified sampling approach does not lead to increased statistical efficiency. (In the two examples of systematic sampling that are given above.. Third. using a stratified sampling approach may be more convenient than aggregating data across groups (though this may potentially be at odds with the previously noted importance of utilizing criterion-relevant strata). its theoretical properties make it difficult to quantify that accuracy. utilizing a stratified sampling method can lead to more efficient statistical estimates (provided that strata are selected based upon relevance to the criterion in question. systematic sampling is an EPS method.24.. much of the potential sampling error is due to variation between neighbouring houses . since each stratum is treated as an independent population. Stratified sampling Where the population embraces a number of distinct categories. in such cases.13.) As described above.14. because all elements have the same probability of selection (in the example given. the set {4..24." Each stratum is then sampled as an independent sub-population. potentially enabling researchers to use the approach best suited (or most cost-effective) for each identified subgroup within the population.e. such a tactic will not result in less efficiency than would simple random sampling..

There are. Implementation usually follows a simple random sample. in some cases (such as designs with a large number of strata. 3. In addition to allowing for stratification on an ancillary variable. as well as leading to increased complexity of population estimates. post stratification can be used to implement weighting. however. the required sample size would be no larger than would be required for simple random sampling. Allows use of different sampling techniques for different subpopulations. identifying strata and implementing such an approach can increase the cost and complexity of sample selection. Disadvantages 1. A stratified sampling approach is most effective when three conditions are met 1. which can improve the precision of a sample's estimates. or those with a specified minimum sample size per group). Can be expensive to implement. This approach is typically implemented due to a lack of prior knowledge of an appropriate stratifying variable or when the experimenter lacks the necessary information to create a stratifying variable during the sampling phase. 4. Variability within strata are minimized 2. First. Although the method is susceptible to the pitfalls of post hoc approaches. Variability between strata are maximized 3. when examining multiple criteria. 2. stratified sampling can potentially require a larger sample than would other methods (although in most cases. 3. Improves the accuracy/efficiency of estimation. Permits greater balancing of statistical power of tests of differences between strata by sampling equal numbers from strata varying widely in size. it can provide several benefits in the right situation. further complicating the design. Requires selection of relevant stratification variables which can be difficult. Is not useful when there are no homogeneous subgroups. but not to others. Focuses on important subpopulations and ignores irrelevant ones. The variables upon which the population is stratified are strongly correlated with the desired dependent variable. stratifying variables may be related to some. some potential drawbacks to using stratified sampling. and potentially reducing the utility of the strata. Finally. Post stratification Stratification is sometimes introduced after the sampling phase in a process called "post stratification". Advantages over other sampling methods 1. Second. . 2.

220. and sixth schools. However. for each element in the population. To do this. and 490 students respectively (total 1500 students). the data are stratified on the target and a sample is taken from each stratum so that the rare target class will be more represented in the sample. The effects of the input variables on the target are often estimated with more precision with the choice-based sample even when a smaller overall sample size is taken. To address this problem. PPS sampling is commonly used for surveys of businesses. the first.Oversampling Choice-based sampling is one of the stratified sampling strategies. the second school 151 to 330 (= 150 + 180). a survey attempting to measure the number of guest-nights spent in hotels might use each hotel's number of rooms as an auxiliary variable. The model is then built on this biased sample. PPS may be combined with a systematic approach. these selection probabilities can then be used as the basis for Poisson sampling.or under-represented due to chance variation in selections. we would select the schools which have been allocated numbers 137. this has the drawback of variable sample size. In choicebased sampling. and different portions of the population may still be over. The results usually must be adjusted to correct for the oversampling. we could allocate the first school numbers 1 to 150. compared to a random sample. 260. and we want to use student population as the basis for a PPS sample of size three. in which the selection probability for each element is set to be proportional to its size measure. 180. as discussed above. The PPS approach can improve accuracy for a given sample size by concentrating sample on large elements that have the greatest impact on population estimates. the third school 331 to 530. 637. . One option is to use the auxiliary variable as a basis for stratification. fourth. If our random start was 137. 200. an older measurement of the variable of interest can be used as an auxiliary variable when attempting to produce more current estimates. Probability proportional to size sampling In some cases the sample designer has access to an "auxiliary variable" or "size measure".for instance. In some cases. up to a maximum of 1.e. In a simple PPS design. and 1137. These data can be used to improve accuracy in sample design. Example: Suppose we have six schools with populations of 150. i. Another option is probability-proportional-to-size ('PPS') sampling. where element size varies greatly and auxiliary information is often available . We then generate a random start between 1 and 500 (equal to 1500/3) and count through the school populations by multiples of 500. and so on to the last school (1011 to 1500). believed to be correlated to the variable of interest.

rather than having to drive to a different block for each household. In the example above. cluster sampling requires a larger sample than SRS to achieve the same level of accuracy . additional samples of units are selected. For this reason. Then judgement is used to select the subjects or units from each segment based on a specified proportion. the sample only requires a block-level city map for initial selections. In following stages. we might choose to select 100 city blocks and then interview every household within the selected blocks. This is a complex form of cluster sampling in which two or more levels of units are embedded one in the other. For example. an interviewer can make a single trip to visit several households in one block. or by time periods. In the example above. depending on how the clusters differ between themselves. if surveying households within a city.although this is rarely taken into account in the analysis. In the second stage. just as in stratified sampling. All ultimate units (individuals. By eliminating the work involved in describing clusters that are not selected.but cost savings from clustering might still make this a cheaper option.) For instance. for instance) selected at the last step of this procedure are then surveyed. Clustering can reduce travel and administrative costs. In quota sampling. the population is first segmented into mutually exclusive sub-groups. This technique. Multistage sampling can substantially reduce sampling costs. clusters can be chosen from a cluster-level frame. Cluster sampling is commonly implemented as multistage sampling. a sample of primary units is randomly selected from each cluster (rather than using all units contained in all selected clusters).Cluster sampling Sometimes it is more cost-effective to select respondents in groups ('clusters'). and then a household-level map of the 100 selected blocks. rather than a household-level map of the whole city. as compared with the within-cluster variation. an interviewer may be told to sample 200 females and 300 males between the age of 45 and 60. and so on. . Instead. (Nearly all samples are in some sense 'clustered' in time . in each of those selected clusters. thus. It also means that one does not need a sampling frame listing all elements in the target population. multistage sampling can reduce the large costs associated with traditional cluster sampling. with an element-level frame created only for the selected clusters. The first stage consists of constructing the clusters that will be used to sample from. is essentially the process of taking random subsamples of preceding random samples. Cluster sampling generally increases the variability of sample estimates above that of simple random sampling. Sampling is often clustered by geography. where the complete population list would need to be constructed (before other sampling methods could be applied).

which would not represent the views of other members of society in such an area. intersects the element. Accidental sampling (sometimes known as grab. Is there good reason to believe that a particular convenience sample would or should respond or behave differently than a random sample from the same population? 3. called a "transect". The researcher using such a sample cannot scientifically make generalizations about the total population from this sample because it would not be representative enough. each participant is given the same survey or interview at two or more time points. Are there controls within the research design or experiment which can serve to lessen the impact of a non-random convenience sample. the people that he/she could interview would be limited to those given there at that given time. It may be through meeting the person or including a person in the sample when one meets them or chosen by finding them through technological means such as the internet or through phone. such as respondent driven sampling. The problem is that these samples may be biased because not everyone gets a chance of selection. thereby ensuring the results will be more representative of the population? 2. Line-intercept sampling is a method of sampling elements in a region whereby an element is sampled if a chosen line segment. where existing study subjects are used to recruit more subjects into the sample. Some variants of snowball sampling. . That is. Is the question being asked by the research one that can adequately be answered using a convenience sample? In social science research. allow calculation of selection probabilities and are probability sampling methods under certain conditions. Panel sampling is the method of first selecting a group of participants through a random sampling method and then asking that group for the same information again several times over a period of time. convenience or opportunity sampling) is a type of nonprobability sampling which involves the sample being drawn from that part of the population which is close to hand. For example interviewers might be tempted to interview those who look most helpful. if the interviewer were to conduct such a survey at a shopping center early in the morning on a given day.It is this second step which makes the technique one of non-probability sampling. This type of sampling is most useful for pilot testing. snowball sampling is a similar technique. a population is selected because it is readily available and convenient. For example. This random element is its greatest weakness and quota versus probability has been a matter of controversy for many years. Several important considerations for researchers using convenience samples include: 1. Therefore. In quota sampling the selection of the sample is non-random. if the survey were to be conducted at different times of day and several times per week.

Ans: 1. this is a WR design. this becomes a WOR design. However.each period of data collection is called a "wave". 8. 4. Replacement of selected units Sampling schemes may be without replacement ('WOR' . measure them. Panel sampling can also be used to inform researchers about within-person health changes due to age or to help explain changes in continuous dependent variables such as spousal interaction. 6. if we eat the fish). Physical appearance of Questionnaire has to be attractive but not such case is necessary with schedule. because we might end up catching and measuring the same fish more than once. Risk of incomplete & wrong information is more in Questionnaire.an element may appear multiple times in the one sample). Respondent should be literate & co-operative in Questionnaire but schedule can be filled by illiterate. Q 4. and structural equation modeling with lagged effects. Distinguish between Schedules and Questionnaires. growth curves. Success of Questionnaire depends on its design but in case of Schedule it depends on honesty & competency of Enumerator Q 5. Questionnaire can be filled by anyone but schedule is always filled by enumerator. Questionnaire can be returned without answering all the questions but. including MANOVA. if we catch fish. Questionnaire can be sent via mail but schedule is done only personally 2. and immediately return them to the water before continuing with the sample. if we do not return the fish to the water (e. 7. for example with regard to chronic illness to job stress to weekly food expenditures. Questionnaire is cheaper method than schedule (for schedule you have to move everywhere 3. This longitudinal samplingmethod allows estimates of changes in the population. There have been several proposed methods of analyzing panel data. What are the problems encountered in the interview? . in schedule.no element can be selected more than once in the same sample) or with replacement ('WR' . enumerator ensures the filling all the questions. 5.g. For example.

such biasing factors can never be overcome completely. They are: · Partial response. but their effects can be reduced by careful selection and training of interviewers. proper motivation and supervision. which arises on account of respondent’s failure to understand a question or lack of information necessary for answering it. in which the respondent’s answer is not relevant to the question asked · Inaccurate response. when the reply is biased or distorted and · Verbalized response problem. standard instructions on probing procedure and so on) and standardization of interviewer behaviour. in which the respondent gives a relevant but incomplete answer · Non-response. Interviewer’s Bias The interviewer is an important cause of response bias. Another source of response bias arises from interviewer’s perception of the situation. As interviewers are human beings. if he regards the assignment as impossible or sees the results of the survey as possible threats to personal interests or beliefs he is likely to introduce bias. non-response and interviewer’s bias. word emphasis. His own attitudes and expectations about what a particular category of respondents may say or think may bias the data. inadequate response. apparent social status. There is need for more research on ways to minimize bias in the interview.Ans: Interview Problems In personal interviewing. Another source of response of the interviewer’s characteristics (education. Inadequate response Kahn and Cannel distinguish five principal symptoms of inadequate response. standardization or interview procedures (use of standard wording in survey questions. tone of voice and question rephrasing. He may resort to cheating by ‘cooking up’ data without actually interviewing. Non-response . when the respondent remains silent or refuses to answer the question · Irrelevant response. the researcher must deal with two major problems. etc) may also bias his answers. The interviewers can influence the responses by inappropriate suggestions.

For example. employed persons may not be available during working hours. refusal. If someone is available. Non-responses reduce the effective sample size and its representativeness. Dispersion: A measure of statistical dispersion is a real number that is zero if all the data are identical. Although. There are many sources of non-response. line respondent’s hours of availability can be ascertained and the next visit can be planned accordingly. then. the measure of dispersion has the same units. such as metres or seconds. Inaccessibility Some respondents may be inaccessible. if the measurements have units. and increases as the data become more diverse. Most measures of dispersion have the same scale as the quantity being measured. This depends upon the nature of the respondent and the time of calls. Non-availability Some respondents may not be available at home at the time of call. Such measures of dispersion include: • • • • • • • Standard deviation Interquartile range or Interdecile range Range Mean difference Median absolute deviation Average absolute deviation (or simply called average deviation) Distance standard deviation . a hardcore of refusals remains. another try or perhaps another approach may find some of them cooperative. non-availability. Selection of appropriate timing for calls could solve this problem. incapacity and inaccessibility. Farmers may not be available at home during cultivation season. Q 6.Non-response refers to failure to obtain responses from some sample respondents. Refusal Some persons may refuse to furnish information because they are ill-disposed. In other words. Incapacity or inability may refer to illness which prevents a response during the entire survey period. Write short notes on the following: a. or approached at the wrong hour and so on. Evenings and weekends may be favourable interviewing hours for such respondents. It cannot be less than zero. This may also arise on account of language barrier. Some may not be found due to migration and other reasons.

equal to twice the Gini coefficient There are other measures of dispersion: • • Variance (the square of the standard deviation) — location-invariant but not linear in scale. they have no units even if the variable itself has units.These are frequently used (together with scale factors) as estimators of scale parameters. Robust measures of scale are those unaffected by a small fraction of outliers. Then. One measure that does so is the discrete entropy. 83. b. The same process works if you have fewer or more numbers in the group. among them the Allan variance and the Hadamard variance. Therefore. These include: • • • Coefficient of variation Quartile coefficient of dispersion Relative mean difference. it is less common to measure dispersion by a single number. Say you earned an 80% and a 90%. you would add the two numbers together and divide the sum by two. 101 and 116. if you were averaging only two numbers. 80% + 90% = 170%. In other words. If you had two grades in a course. 99. your current average is 85%. 475 divided by 5 = 95. To arrive at the average score of this group of numbers. 170% divided by 2 = 85%. Some measures of dispersion have specialized purposes. as count data are themselves dimensionless: otherwise this is not scalefree. a list of bowling scores might be: 76. Mathematical averages: To arrive at a mathematical average. usually you start with a list of numbers. For categorical variables. Other measures of dispersion are dimensionless (scale-free). Variance-to-mean ratio — mostly used for count data when the term coefficient of dispersion is used and when this ratio is dimensionless. All the above measures of statistical dispersion have the useful property that they are location-invariant. . you would first add the grades together. So if a random variable X has a dispersion of SX then a linear transformation Y = aX + b for real a and b should have dispersion SY = |a|SX. divide the sum total by the total number of scores in the group. as well as linear in scale. 76 + 83 + 99 + 101 +116 = 475. the average number of the bowling scores is 95. Therefore. For instance. See qualitative variation. Then divide the sum total by two. in which capacity they are called estimates of scale. For instance. simply add the scores all together.

Sign up to vote on this title
UsefulNot useful