MB0050 Research Methodology Semester: 3 - Assignment Set: 1 1. a.

Differentiate between nominal, ordinal, interval and ratio scales, wi th an example of each. Answer. The "levels of measurement", or scales of measure are expressions that typically refer to the theory of scale types developed by the psychologist Stanl ey Smith Stevens. Stevens proposed his theory in a 1946 Science article titled " On the theory of scales of measurement" In that article, Stevens claimed that al l measurement in science was conducted using four different types of scales that he called "nominal", "ordinal", "interval" and "ratio". The theory of scale types Stevens (1946, 1951) proposed that measurements can be classified into four diff erent types of scales. These are shown in the table below as: nominal, ordinal, interval, and ratio. Scale Type Permissible Statistics Admissible Scale Transformation Mathemat ical structure nominal (also denoted as categorical) mode, Chi-square One to One (equa lity (=)) standard set structure (unordered) ordinal median, percentile Monotonic increasing (order (<)) totally ordered set interval mean, standard deviation, correlation, regression, analysis of v ariance Positive linear (affine) affine line ratio All statistics permitted for interval scales plus the following: geometr ic mean, harmonic mean, coefficient of variation, logarithms Positive similar ities (multiplication) field Nominal scale At the nominal scale, i.e., for a nominal category, one uses labels; for example , roc s can be generally categorized as igneous, sedimentary and metamorphic. Fo r this scale, some valid operations are equivalence and set membership. Nominal measures offer names or labels for certain characteristics. Variables assessed on a nominal scale are called categorical variables; see also categorical data. Stevens (1946, p. 679) must have nown that claiming nominal scales to measure o bviously non-quantitative things would have attracted criticism, so he invo ed h is theory of measurement to justify nominal scales as measurement: the use of numerals as names for classes is an example of the assignment of numer als according to rule. The rule is: Do not assign the same numeral to different classes or different numerals to the same class. Beyond that, anything goes with the nominal scale. The central tendency of a nominal attribute is given by its mode; neither the me an nor the median can be defined. We can use a simple example of a nominal category: first names. Loo ing at nearb y people, we might find one or more of them named Aamir. Aamir is their label; a nd the set of all first names is a nominal scale. We can only chec whether two people have the same name (equivalence) or whether a given name is in on a certa in list of names (set membership), but it is impossible to say which name is gre ater or less than another (comparison) or to measure the difference between two names. Given a set of people, we can describe the set by its most common name (t he mode), but cannot provide an "average name" or even the "middle name" among a ll the names. However, if we decide to sort our names alphabetically (or to sort them by length; or by how many times they appear in the US Census), we will beg in to turn this nominal scale into an ordinal scale. Ordinal scale Ran -ordering data simply puts the data on an ordinal scale. Ordinal measurement s describe order, but not relative size or degree of difference between the item s measured. In this scale type, the numbers assigned to objects or events repres

ent the ran order (1st, 2nd, 3rd, etc.) of the entities assessed. A Li ert Scal e is a type of ordinal scale and may also use names with an order such as: "bad" , "medium", and "good"; or "very satisfied", "satisfied", "neutral", "unsatisfie d", "very unsatisfied." An example of an ordinal scale is the result of a horse race, which says only which horses arrived first, second, or third but include n o information about race times. Another is the Mohs scale of mineral hardness, w hich characterizes the hardness of various minerals through the ability of a har der material to scratch a softer one, saying nothing about the actual hardness o f any of them. Yet another example is military ran s; they have an order, but no well-defined numerical difference between ran s. When using an ordinal scale, the central tendency of a group of items can be des cribed by using the group's mode (or most common item) or its median (the middle -ran ed item), but the mean (or average) cannot be defined. In 1946, Stevens observed that psychological measurement usually operates on ord inal scales, and that ordinary statistics li e means and standard deviations do not have valid interpretations. Nevertheless, such statistics can often be used to generate fruitful information, with the caveat that caution should be ta en i n drawing conclusion from such statistical data. Psychometricians li e to theorise that psychometric tests produce interval scale measures of cognitive abilities (e.g. Lord & Novic , 1968; von Eye, 2005) but t here is little prima facie evidence to suggest that such attributes are anything more than ordinal for most psychological data (Cliff, 1996; Cliff & Keats, 2003 ; Michell, 2008). In particular IQ scores reflect an ordinal scale, in which all scores are only meaningful for comparison, rather than an interval scale, in wh ich a given number of IQ "points" corresponds to a unit of intelligence Thus it is an error to write that an IQ of 160 is just as different from an IQ of 130 as an IQ of 100 is different from an IQ of 70. In mathematical order theory, an ordinal scale defines a total preorder of objec ts (in essence, a way of sorting all the objects, in which some may be tied). Th e scale values themselves (such as labels li e "great", "good", and "bad"; 1st, 2nd, and 3rd) have a total order, where they may be sorted into a single line wi th no ambiguities. If numbers are used to define the scale, they remain correct even if they are transformed by any monotonically increasing function. This prop erty is nown as the order isomorphism. A simple example follows: Judge's score x Score minus 8 x-8 Tripled score 3x Cubed score x3 Alice's coo ing ability 10 2 30 1000 Bob's coo ing ability 9 1 27 729 Claire's coo ing ability 8.5 0.5 25.5 614.125 Dana's coo ing ability 8 0 24 512 Edgar's coo ing ability 5 -3 15 125 Since x-8, 3x, and x3 are all monotonically increasing functions, replacing the ordinal judge's score by any of these alternate scores does not affect the relat ive ran ing of the five people's coo ing abilities. Each column of numbers is an equally legitimate ordinal scale for describing their abilities. However, the n umerical (additive) difference between the various ordinal scores has no particu lar meaning. Interval scale Quantitative attributes are all measurable on interval scales, as any difference between the levels of an attribute can be multiplied by any real number to exce ed or equal another difference. A highly familiar example of interval scale meas urement is temperature with the Celsius scale. In this particular scale, the uni t of measurement is 1/100 of the temperature difference between the freezing and

boiling points of water under a pressure of 1 atmosphere. The "zero point" on a n interval scale is arbitrary; and negative values can be used. The formal mathe matical term is an affine space (in this case an affine line). Variables measure d at the interval level are called "interval variables" or sometimes "scaled var iables" as they have units of measurement. Ratios between numbers on the scale are not meaningful, so operations such as mu ltiplication and division cannot be carried out directly. But ratios of differen ces can be expressed; for example, one difference can be twice another. The central tendency of a variable measured at the interval level can be represe nted by its mode, its median, or its arithmetic mean. Statistical dispersion can be measured in most of the usual ways, which just involved differences or avera ging, such as range, interquartile range, and standard deviation. Since one cann ot divide, one cannot define measures that require a ratio, such as studentized range or coefficient of variation. More subtly, while one can define moments abo ut the origin, only central moments are useful, since the choice of origin is ar bitrary and not meaningful. One can define standardized moments, since ratios of differences are meaningful, but one cannot define coefficient of variation, sin ce the mean is a moment about the origin, unli e the standard deviation, which i s (the square root of) a central moment. b. What are the purposes of measurement in social science research? Answer. Three Purposes of Research Social research can serve a variety of purposes. Three of the most influential a nd common purposes of research are exploration, description and explanation. Exploration involves familiarizing a researcher with a topic. Exploration satisf ies the researcher's curiosity and desire for improved understanding. Exploratio n tests the feasibility of underta ing a more extensive study. Exploration helps develop the methods that will be used in a study. Description involves describing situations and events through scientific observa tion. Scientific descriptions are typically more accurate and precise than causa l ones. For example, the U. S. Census uses descriptive social research in its ex amination of characteristics of the U. S. population. Explanation involves answering the questions of what, where, when, and how. Expl anatory studies answer questions of why. For example, an explanatory analysis of the 2002 General Social Survey (GSS) data indicates that 38 percent of men and 30 percent of women said marijuana should be legalized, while 55 percent of libe rals and 27 percent of conservatives said the same. Given these statistics, you could start to develop an explanation for attitudes toward marijuana legalizatio n. In addition, further study of gender and political orientation could lead to a deeper explanation of this issue. The Logic of Idiographic vs. Nomothetic Explanation Idiographic explanation - a "full", detailed, in-depth understanding of a case; for practical reasons, only a few subjects are studied in this way. An idiograph ic explanation of the marijuana legalization survey would involve a more conclus ive list of factors that could influence a person's viewpoints on this issue. Th erefore, an idiographic explanation would need to consider several factors, such as information from parents and previous experiences, not just political orient ation. Nomothetic explanation - a generalized understanding of a given case, with the g oal of finding new factors that can account for many of the variations in a give n phenomenon; is applicable to many subjects. Regarding the survey mentioned abo ve dealing with people's stances on marijuana legalization, a nomothetic explana tion may simply suggest that political orientation is the main driving force beh ind people's differing opinions on this issue. Hypotheses are not required in no mothetic research. o There are three main criteria for nomothetic causal relationships in soc ial research: 1) the variables must be correlated 2) the variables are nonspurious

3) the cause ta es place before the effect Correlation - an empirical relationship between two variables such that changes in one are associated with changes in the other, or particular attributes in one are associated with particular attributes in the other. Spurious relationship - a coincidental statistical correlation between two varia bles shown to be caused by some third variable. For example, increased ice cream consumption is related to the crime rate rise. But this relationship is caused by a third variable, summertime yielding hot weather and closed schools. Therefo re, for a causal relationship, variables must be nonspurious False criteria for nomothetic causality: Complete causation - proper nomothetic explanation is probabilistic and does not explain every single case. Exceptional cases - exceptions do not disprove nomothetic explanation. Majority of cases - nomothetic explanation may be applicable to only a minority of cases in a given situation. Necessary and Sufficient Causes: A necessary cause represents a condition that must be present for the effect to follow. Example: It is necessary for you to ta e college courses in order to get a degree. Ta e away the courses, and the degree never follows. ***A sufficient cause represents a condition that guarantees the effect if it is present. Example: S ipping an exam would be a sufficient cause for failing it ( even though there are other ways to fail it). 2. a. What are the sources from which one may be able to identify research problems? Answer. SOURCES OF PROBLEMS 1. Curiosity One of the oldest and most common sources of research problems is curiosity. Jus t as your interest in baseball or gardening may stimulate you to investigate the topic in greater depth, so researchers may investigate phenomena that attract t heir personal interest. For example, a man named Lipset belonged to the Internat ional Typographical Union, and his son's well- nown study of democratic decision -ma ing processes was specifically concerned with that organization. The younger Lipset was curious because the union was very large and, according to contempor ary social thin ing, should have succumbed to an oligarchic decision-ma ing proc ess (a rule of the many by the few), yet appeared to be very democratic. 2. Significant Others While personal interests or curiosity may be ey motivating factors in problem s election, it's a basic assumption in the social sciences that individuals learn from and are influenced by others. Because researchers are usually recruited and trained in universities, the selection of research topics often reflects the in fluence of teachers or fellow students. For example, the first boo of Carlos Ca staneda's adventures with Don Juan (The Teachings of Don Juan) served as his mas ter's thesis, while the third (Journey to Ixtlan) was his Ph.D. dissertation in anthropology. A reading of his advisor's writings suggests that the character of Don Juan is based, at least in part, on Castaneda's advisor, Harold Garfin el. Garfin el's notion that people actually construct social reality instead of mere ly reacting to it is a major theme in Castaneda's wor s. This type of influence may also be les personal, as when one is influenced by the writings of past theo rists. 3. Social Problems In addition to personal curiosity and the influence of others, concern with soci al problems has been a major source of social research. As discussed in more det ail in Chapter 4, one of the first uses of social surveys was to study poverty. Li ewise, concerns with the Nazi extermination of 6 million Jews and discriminat ion in this country have generated a vast body of psychological and social-psych ological research on ethnic, gender, sexual orientation, and racial prejudice. C oncern with the energy crisis and the threat of overpopulation have generated vo lumes of research. Just as researchers' interests vary, so they also differ in what they perceive t o be a social problem. Karl Marx's view of capitalism as a social problem is not

shared by all social scientists. Li ewise one may view the feminist movement as a social problem, or one might see sexual orientation stereotyping and prejudic e to be problems. Still, this interest in social problems suggests that research ers are concerned not only with nowledge for nowledge's sa e, but also with it s application, or with what in Chapter 1 we called intelligent intervention. Regardless of precipitating factors, one of the basic goals of research is to in crease our nowledge about a particular phenomenon. For the scientist this means building a body of theory. As you read in the previous chapter, nowing is a ma tter of providing a description of reality. While we could conceivably call any description of reality a "theory," the term is generally used more rigorously. W hen a statement that lin s together two or more concepts is widely agreed upon, it is given the status of a proposition. A theory consists of a set of propositi ons that are systematically interrelated and purport to explain some phenomenon. Probably most research is conducted in an effort to test propositions. The resea rch begins with a theory and uses empirical evidence to find out how well that d escription actually fits reality. If amenable to observation, either the basic p ropositions or deduced propositions may be tested against empirical data. In tes ting theory, the goal is not merely to find support for one theory, but also to eliminate less li ely theories. Thus, Dur heim began his study of suicide by att empting to show that previous theories of suicide could not be true. Dur heim fi rst tried to demonstrate that suicide is not a result of mental illness or psych opathic states by showing that rates of suicide and insanity are not related. In the same manner he attempted to discredit theories that attributed the causes o f suicide to race, heredity, climate, seasonal temperatures, and imitation befor e defending his theory of suicide in terms of social forces. Research also has the dual role of both specifying and generalizing the domain o f the theory. For example, a theory about social power may state that power unde rlies all organizational relationships. A more specific theoretical statement co uld be that while people at all organizational levels hold certain inds of powe r, those in higher-status positions always have more power than those in lower-s tatus positions. One might also attempt to expand the relationship between power and organizational hierarchies to include other social hierarchies. In this cas e it might be stated that lawyers have more power than clients or doctors have m ore power than patients. Naturally, specification and generalization of theory a re major concerns because we strive to explain as many phenomena as possible. b. Why literature survey is important in research? Ans. Research is made in order to inform people with new nowledge or discovery. However, it is not to be expected that everybody would willingly believe what y ou are tac ling in your whole research paper. Thus, what you can do to ma e your research more credible will be to support them with other wor s which have spo en about the same topic that you have for your research. This is where literatur e review comes in. You can even have literature sources in wor s such as stories, comments, project , speech, article, novel, poem, essay, program, theory, and others. This is why literature review involves scanning the pages of any published literature li e b oo s, newspaper, magazine, website, webpage, collection, paper, pamphlet, and th e li e where you may be able to find any reference to the same topic that you ar e researching on. This time, literature does not exclusively refer to the poetic re ndition of words, li e that of Sha espeare alone. There are many reasons why literature review is rendered as a significant part o f any research or dissertation paper. You may as what ma es it as such if it is only supposed to contain tidbits of other related wor s. Literature review is t he part of the paper where the researcher will be given the opportunity to stren gthen your paper for you will be citing what other reliable authors have said ab out your topic. This will prove that you are not just writing about any random s ubject but that many others have also poured their thoughts on the topic.

You may also as what ma es literature review a necessary part of the paper. Thi s question can be answered by trying not to include the review in your paper. Ob viously, it affects the length of your paper but this is not the noticeable part . What would most certainly be lac ing is the fact that your paper, without the literature review, only contains all of your opinions about the facts that you h ave discovered through your research. Thus, how can you further convince the rea ders, in this case, the committee who will scrutinize your paper? This is the ne ed that is answered only by the literature review. By the mere fact that you are using referencing by citing what more credible people had said about the topic wi ll build a stronger foundation for your paper. With a literature review, you need to establish a clear tie between the wor s th at you have cited and the topic that you are writing about. You should be able t o justify the inclusion of a certain wor in your review so as to ma e everythin g that you have written useful. The more you include useless points in your pape r, the more that the committee will thin that you have not put in a lot of thin ing into your paper. Literature review is also unique from the rest of the paper. While you have to f ill most of the paper with your own analysis, in a literature review alone, you will have to write purely about related wor s of other people. 3. a. What are the characteristics of a good research design? Ans. Characteristics of Research Design Generally a good research design minimizes bias and maximizes the reliability of the data collected and analyzed. The design which gives the smallest experimen tal error is reported to be the best design in scientific investigation. Simila rly, a design which yields maximum information and provides a opportunity for co nsidering different aspects of a problem is considered to be the most appropriat e efficient design . Thus the question of a good design is related to the purpo se or objective of the research problem and also with the nature of the problem to be studied. A good research design should satisfy the following four conditi ons namely objectivity, reliability, validity and generalization of the findings . 1. Objectivity: It refers to the findings related to the method of data collect ion and scoring of the responses. The research design should permit the measuri ng instrument which are fairly objective in which every observer or judge scorin g the performance must precisely give the same report. In other words, the obje ctivity of the procedure may be judged by the degree of agreement between the fi nal scores assigned to different individuals by more than one independent observ er. This ensures the objectivity of the collected data which shall be capable o f analysis and drawing generalizations. 2. Reliability: Reliability refers to consistency through out a series of measu rements. For eg: if a respondent gives out a response to a particular item, he is expected to give the same response to that item even if he is as ed repeatedl y. If he is changing his response to the same item, the consistency will be los t. So the researcher should frame the items in a questionnaire in such a way th at it provides consistency or reliability. 3. Validity: Any measuring device or instrument is said to be valid when it meas ures what it is expected to measure. For eg: an intelligence test concucted for measuring the I.Q should measure only the intelligence and nothing else, and th e questionnaire shall be framed accordingly. 4. Generalizability: It means how best the data collected from the samples can be utilized for drawing certain generalisations applicable to a large group from which sample is drawn. Thus a research design helps an investigator to general ize his findings provided he has ta en due care in defining the population, sele

cting the sample, deriving appropriate statistical analysis etc. while preparin g the research design. Thus a good research design is one which is methodologic ally prepared and should ensure that: a) The measuring instrument can yield objective, reliable and valid data. b) The population is clearly defined. c) Most appropriate techniques of sample selection is used to form an appropriat e sample. d) Appropriate statical analysis has been carried out, and e) The findings of the study is capable of generalisations. b. What are the components of a research design? Ans. Wor ing Design - preliminary plan for beginning a qualitative research proj ect. ! ! ! !

purposeful sampling - nonrandom sampling design that selects subjects or sites due to specific characteristics or phenomena under study Wor ing Hypotheses - hypotheses regarding possible outcomes that guide the resea rch study (also see foreshadowed problems) Data Collection - forms of data collection vary widely depending on the qualitat ive research design employed. ! the researcher must be able to gain access to the situation ! the researcher must decide on his/her role in data collection ! participant-observer or observer only ! interactive or noninteractive data collection ! the researcher must decide upon the format for data collection ! interviews ! observations ! oral histories ! specimen records ! document collection & review ! tape recording &/or video recording ! the researcher must employ some system for managing great volume of data collected ! reflective journal/log - record of events and personal thoughts as data collection proceeds--may provide basis for changes in wor ing design or reformulation of hy potheses Specimen record - a narrative description of one person in a natural sit uation as seen by s illed observers over a substantial period of time. Record stream of behavior

divide stream into units

Oral History - interviews conducted with the use of a tape recorder. ! ! ! ! entire conversation recorded allows examination of inflections emphasize open-ended questions analyzed through listening to tape rather than transcribing it

Data Analysis and Interpretation

subjects to be studied sites to be studied time frame for data collection possible variables to be considered

analyze units

Data Analysis - begins soon after data collection begins to allow verification, refinement, or restatement of wor ing hypotheses/ foreshadowed problems/or initi al theories. | qualitative data analysis is a series of successive approximations towar d an accurate description and interpretation of phenomena under study Coding - process of data reduction through data organization--allows researche r to "see what's there" for purposes of: ! sorting/categorization ! comparison with initial hypotheses/problem statements (support or contradict earlier ideas) ! refinement of design/hypotheses Characteristics of a coding system: ! able to accurately capture information relevant to the research problem ! information captured is useful in describing and understanding the phenom enon being studied Coding categories: ! are influenced by the purpose and context of the study ! are specific to the study ! may be determined before or after data collection and review ! need not be mutually exclusive (e.g. student & teacher perceptions/student demographic characteristics/etc) Possible Codes: ! Setting/Context Codes - reflect the context or setting in which phenomena under study occur (e.g. several different settings-high school/vocational s chool/etc.included under school environment) ! Process Codes - focus on the sequence of events and how changes occur dur ing the study (e.g. different ways in which students go about dropping ou t of school)

4. a. Distinguish between Doubles sampling and multiphase sampling. Ans. Double Sampling Plans:

How double sampling plans wor ? Double and multiple sampling plans were invented to give a questionable lot anot her chance. For example, if in double sampling the results of the first sample a re not conclusive with regard to accepting or rejecting, a second sample is ta e n. Application of double sampling requires that a first sample of size n1 is ta en at random from the (large) lot. The number of defectives is then counted and compared to the first sample's acceptance number a1 and rejection number r1. Den ote the number of defectives in sample 1 by d1 and in sample 2 by d2, then: If d1 a1, the lot is accepted. If d1 r1, the lot is rejected. If a1 < d1 < r1, a second sample is ta en. If a second sample of size n2 is ta en, the number of defectives, d2, is counted

. The total number of defectives is D2 = d1 + d2. Now this is compared to the ac ceptance number a2 and the rejection number r2 of sample 2. In double sampling, r2 = a2 + 1 to ensure a decision on the sample. If D2 a2, the lot is accepted. If D2 r2, the lot is rejected. Design of a double sampling plan? The parameters required to construct the OC curve are similar to the single samp le case. The two points of interest are (p1, 1- ) and (p2, , where p1 is the lo t fraction defective for plan 1 and p2 is the lot fraction defective for plan 2. As far as the respective sample sizes are concerned, the second sample size mus t be equal to, or an even multiple of, the first sample size. There exist a variety of tables that assist the user in constructing double and multiple sampling plans. The index to these tables is the p2/p1 ratio, where p2 > p1. One set of tables, ta en from the Army Chemical Corps Engineering Agency f or = .05 and = .10, is given below: Tables for n1 = n2 accept approximation values R = numbers of pn1 for p2/p1 c1 c2 P = .95 P = .10 ________________________________________ 11.90 0 1 0.21 2.50 7.54 1 2 0.52 3.92 6.79 0 2 0.43 2.96 5.39 1 3 0.76 4.11 4.65 2 4 1.16 5.39 4.25 1 4 1.04 4.42 3.88 2 5 1.43 5.55 3.63 3 6 1.87 6.78 3.38 2 6 1.72 5.82 3.21 3 7 2.15 6.91 3.09 4 8 2.62 8.10 2.85 4 9 2.90 8.26 2.60 5 11 3.68 9.56 2.44 5 12 4.00 9.77 2.32 5 13 4.35 10.08 2.22 5 14 4.70 10.45 2.12 5 16 5.39 11.41 Multiple sampling: Multiple Sampling is an extension of the double sampling concept Multiple sampling is an extension of double sampling. It involves inspection of 1 to successive samples as required to reach an ultimate decision. Mil-Std 105D suggests = 7 is a good number. Multiple sampling plans are usuall y presented in tabular form: Procedure for multiple sampling? The procedure commences with ta ing a random sample of size n1from a large lot o f size N and counting the number of defectives, d1. if d1 a1 the lot is accepted. if d1 r1 the lot is rejected. if a1 < d1 < r1, another sample is ta en. If subsequent samples are required, the first sample procedure is repeated sampl e by sample. For each sample, the total number of defectives found at any stage, say stage i, is This is compared with the acceptance number ai and the rejection number ri for t hat stage until a decision is made. Sometimes acceptance is not allowed at the e arly stages of multiple sampling; however, rejection can occur at any stage. Efficiency measured by the ASN

Efficiency for a multiple sampling scheme is measured by the average sample numb er (ASN) required for a given Type I and Type II set of errors. The number of sa mples needed when following a multiple sampling scheme may vary from trial to tr ial, and the ASN represents the average of what might happen over many trials wi th a fixed incoming defect level. b. What is replicated or interpenetrating sampling? The experiment should be reaped more than once. Thus, each treatment is applied in many experimental units instead of one. By doing so, the statistical accura cy of the experiments is increased. For example, suppose we are to examine the effect of two varieties of rice. For this purpose we may divide the field into two parts and grow one variety in one part and the other variety in the other pa rt. We can compare the yield of the two parts and draw conclusion on that basis . But if we are to apply the principle of replication to this experiment, then we first divide the field into several parts, grow one variety in half of these parts and the other variety in the remaining parts. We can collect the data yie ld of the two varieities and draw conclusion by comaring the same. The result so obtained will be more reliable in comparison to the conclusion we draw without applying the principle of replication. The entire experiment can even by repeated several times for better results. Co nsequently replication does not present any difficulty, but computationally it d oes. However, it should be remembered that replication is introduced in order t o increase the precision of a study, that is to say, to increase the accuracy wi th which the main effects and interations can be estimated. 5. a. How is secondary data useful to researcher? Secondary data is information gathered for purposes other than the completion of a research project. A variety of secondary information sources is available to the researcher gather ing data on an industry, potential product applications and the mar et place. Secondary data is also used to gain initial insight into the research problem. Secondary data is classified in terms of its source either internal or external. Internal, or in-house data, is secondary information acquired within the organi zation where research is being carried out. External secondary data is obtained from outside sources. The two major advantages of using secondary data in mar et research are time and cost savings. The secondary research process can be completed rapidly generally in 2 to 3 wee . Substantial useful secondary data can be collected in a matter of days by a s illful analyst. When secondary data is available, the researcher need only locate the source of the data and extract the required information. Secondary research is generally less expensive than primary research. The bul o f secondary research data gathering does not require the use of expensive, speci alized, highly trained personnel. Secondary research expenses are incurred by the originator of the information. There are also a number of disadvantages of using secondary data. These include: Secondary information pertinent to the research topic is either not available, o r is only available in insufficient quantities. Some secondary data may be of questionable accuracy and reliability. Even govern ment publications and trade magazines statistics can be misleading. For example, many trade magazines survey their members to derive estimates of mar et size, m ar et growth rate and purchasing patterns, then average out these results. Often

these statistics are merely average opinions based on less than 10% of their me mbers. Data may be in a different format or units than is required by the researcher. Much secondary data is several years old and may not reflect the current mar et conditions. Trade journals and other publications often accept articles six mont hs before appear in print. The research may have been done months or even years earlier. As a general rule, a thorough research of the secondary data should be underta e n prior to conducting primary research. The secondary information will provide a useful bac ground and will identify ey questions and issues that will need to be addressed by the primary research. Internal data sources Internal secondary data is usually an inexpensive information source for the com pany conducting research, and is the place to start for existing operations. Int ernally generated sales and pricing data can be used as a research source. The u se of this data is to define the competitive position of the firm, an evaluation of a mar eting strategy the firm has used in the past, or gaining a better unde rstanding of the companys best customers. There are three main sources of internal data. These are: 1. Sales and mar eting reports. These can include such things as: Type of product/service purchased Type of end-user/industry segment Method of payment Product or product line Sales territory Salesperson Date of purchase Amount of purchase Price Application by product Location of end-user 2. Accounting and financial records. These are often an overloo ed source of int ernal secondary information and can be invaluable in the identification, clarifi cation and prediction of certain problems. Accounting records can be used to eva luate the success of various mar eting strategies such as revenues from a direct mar eting campaign. There are several problems in using accounting and financial data. One is the ti meliness factor it is often several months before accounting statements are avai lable. Another is the structure of the records themselves. Most firms do not ade quately setup their accounts to provide the types of answers to research questio ns that they need. For example, the account systems should capture project/produ ct costs in order to identify the companys most profitable (and least profitable) activities. Companies should also consider establishing performance indicators based on fina ncial data. These can be industry standards or unique ones designed to measure ey performance factors that will enable the firm to monitor its performance over a period of time and compare it to its competitors. Some example may be sales p er employee, sales per square foot, expenses per employee (salesperson, etc.). 3. Miscellaneous reports. These can include such things as inventory reports, se rvice calls, number (qualifications and compensation) of staff, production and R &D reports. Also the companys business plan and customer calls (complaints) log c an be useful sources of information. External data sources There is a wealth of statistical and research data available today. Some sources are: Federal government Provincial/state governments Statistics agencies Trade associations General business publications

 Magazine and newspaper articles Annual reports Academic publications Library sources Computerized bibliographies Syndicated services. A good place to start your search is the local city, college or university libra ry. Most reference librarians are very nowledgeable about what data is availabl e, or where to loo to find it. Also contact government libraries and department s for research reports/publications they may have done. b. What are the criteria used for evaluation of secondary data? When a researcher wants to use secodary data for his research, he should evaluat e them before deciding to use them. 1.Data Pertinence: The first consideration in evaluation is to examine the pertinence of the available secondary data to the research problem under study. The following questions should be considered. a) What are the definitions and classifications employed? Are they consistent? b) What are the measurements of variables used? What is the degree to which the y conform to the requirements of our research? c What is the coverage of the secondary data in terms of topic and time? Does this coverage fit the needs of our research? 2.Data Quality: If the researcher is convinced about the available secondary dat a for his needs, the next step is to examine the quality of the data. The quali ty of data refers to their accuracy, reliability and completeness. The assuranc e and reliability of the available secondary data depends on the organization wh ich collected them and the purpose for which they were collected. What is the a uthority and prestige of the organization? Is it well recognized? Is it noted for reliability? It is capable of collecting reliable data? Does it use traine d and well qualified investigators? The answers to these questions determines t he degree of confidence we can have in the data and their accuracy. It is impor tant to go to the original source of the secondary data rather than to use an im mediate source which has quoted from the original. Then only, the researcher can review the cautionary ands other comments that wer e made in the original source. 3.=Data Completeness: The completeness refers to the actual converage of the p ublished data. This depends on the methodology and sampling design adopted by t he original organization. Is the methodology sound? Is the sample sixe samll o r large? Is the sampling method appropriate? Answers to these questions may in dicate the appropriateness and adequacy of the data for the problem under study. The question of possible bias should also be examined. Whether the purpose fo r which the original organization collected the data had a particular orientatio n? Has the study been made to promote the organizations interest? How the study was conducted? These are important clues. The researcher must be on guard whe n the source does not report the methodology and sampling design. Then it is no t possible to determine the adequacy of the secondary data for the researchers st udy. 6. What are the differences between observation and interviewing as methods of data collection? Give two specific examples of situations where either observation or interviewing would be more appropriate. Interviews

In interviews information is obtained through inquiry and recorded by enumerator s. Structured interviews are performed by using survey forms, whereas open inter views are notes ta en while tal ing with respondents. The notes are subsequently structured (interpreted) for further analysis. Open-ended interviews, which nee d to be interpreted and analysed even during the interview, have to be carried o ut by well-trained observers and/or enumerators. As in preparing a questionnaire, it is important to pilot test forms designed fo r the interviews. The best attempt to clarify and focus by the designer cannot a nticipate all possible respondent interpretations. A small-scale test prior to a ctual use for data collection will assure better data and avoid wasting time and money. Although structured interviews can be used to obtain almost any information, as with questionnaires, information is based on personal opinion. Data on variables such as catch or effort are potentially subject to large errors, due to poor es timates or intentional errors of sensitive information. Open-ended interviews Open-ended interviews cover a variety of data-gathering activities, including a number of social science research methods. Focus groups are small (5-15 individuals) and composed of representative members of a group whose beliefs, practises or opinions are sought. By as ing initial q uestions and structuring the subsequent discussion, the facilitator/interviewer can obtain, for example, information on common gear use practices, responses to management regulations or opinions about fishing. Panel surveys involve the random selection of a small number of representative i ndividuals from a group, who agree to be available over an extended period - oft en one to three years. During that period, they serve as a stratified random sam ple of people from whom data can be elicited on a variety of topics. Structured interview Generally, structured interviews are conducted with a well-designed form already established. Forms are filled in by researchers, instead of respondents, and in that it differs from questionnaires. While this approach is more expensive, mor e complicated questions can be as ed and data can be validated as it is collecte d, improving data quality. Interviews can be underta en with variety of data sou rces (fishers to consumers), and through alternative media, such as by telephone or in person. Structured interviews form the basis for much of the data collection in small-sc ale fisheries. In an interview approach for sample catch, effort and prices, the enumerators wo r according to a schedule of landing site visits to record data. Enumerators ca n be mobile (that is sites are visited on a rotational basis) or resident at a s pecific sampling site. Their job is to sample vessels, obtaining data on landing s, effort and prices from all boat/gear types that are expected to operate durin g the sampling day. The sample should be as representative as possible of fleet activities. Some additional data related to fishing operations may be required f or certain types of fishing units, such as beach seines or boats ma ing multiple fishing trips in one day. For these, the interview may cover planned activities as well as activities already completed. In an interview approach for boat/gear activities, the enumerators wor accordin g to a schedule of homeport visits to record data on boat/gear activities. Enume rators can be mobile (that is homeports are visited on a rotational basis) or re sident at a specific sampling site. In either case, their job is to determine th e total number of fishing units (and if feasible, fishing gears) for all boat/ge ar types based at that homeport and number of those that have been fishing durin g the sampling day. There are several ways of recording boat/gear activities. In many cases, they co mbine the interview method with direct observations. Direct observations can be used to identify inactive fishing units by observing those that are moored or be ached, and the total number of vessels based at the homeport are already nown, perhaps from a frame survey or register. Often enumerators will still have to ve

rify that vessels are fishing as opposed to other activities by using interviews during the visit. The pure interview approach can be used in those cases where a pre-determined su b-set of the fishing units has been selected. The enumerator's job is to trace a ll fishers on the list and, by means of interviewing, find out those that had be en active during the sampling day. For sites involving a wor able number of fish ing units (e.g. not larger than 20), the interview may involve all fishing units . Sometimes it is possible to as questions on fishing activity which refer to the previous day or even to two days bac . This extra information increases the sam ple size significantly with little extra cost, ultimately resulting in better es timates of total fishing effort. Experience has shown that most of the variabili ty in boat/gear activity is in time rather than space. Direct observations Observers Observers can ma e direct measurements on the fishing vessels, at landing sites, processing plants, or in mar ets. The variables that enumerators can collect in clude catch (landing and discards), effort, vessel/gears, operations, environmen tal variables (e.g. sea state, temperature), biological variables (e.g. length, weight, age), the values and quantities of landings and sales. In practice, observers do not only ma e direct measurements (observations), but also conduct interviews and surveys using questionnaires. They might also be inv olved in data processing and analysis. The tas s of an observer are difficult an d adequate training and supervision are therefore essential. Clear decisions need to be made on the nature and extent of data collected durin g any one trip. Often, the amount of data and frequency of collection can be est ablished analytically with preliminary data. Preferably, observers should only collect data, not carry out other activities, such as enforcement, licensing or tax collection. This should help to minimise b ias by reducing the incentives to lie. Problems in terms of conflicts between da ta collection and law enforcement, for example, can be reduced by clear demarcat ion, separating activities by location or time. This becomes a necessity for atsea observers. Their positions on fishing vessels and the tas s that they perfor m depend significantly on a good wor ing relationship with the captain and crew, which can be lost if they are perceived as enforcement personnel. The major data obtained through at-sea observers are catch and effort data, whic h are often used for cross chec ing fishing logs. At the same time, the at-sea o bservers can collect extra biological (fish size, maturity, and sex), by-catch a nd environmental data, as well as other information on the gears, fishing operat ions etc. Frequently, discards data can only be collected by at-sea observers. The main data obtained from observers at landing sites, processing plants and ma r ets include landing (amount, quality, value and price), biological (size, matu rity), and effort (how many hauls, hours fishing) data. For the large-scale fish ery where a logboo system is used, data collected at landing sites could be use d to crosschec data recorded in logboo s. Data collected from processing plants include quantities by species and, especially in modern factory practices, the batch number of raw materials, which can sometimes be traced bac to fishing ves sels. These data if collected can be used to validate landing data. Collecting data to estimate raising factors for converting landed processed fish weight to the whole weight equivalent may be necessary. By sampling fish before and after processing, conversion factors may be improved. Potential seasonal, l ife history stage and other variations in body/gut weight ratios suggest date, s pecies, sex and size should be recorded in samples. Economic and demographic data at each level (e.g. input and output of various pr oducts to and from mar et and processors) are usually obtained by interview and questionnaire. However, the data directly collected by enumerators can also be t he major source as well as supporting data for those collected through other met hods. While product data in processing plants can be collected through questionnaire ( 6.3.2) or interview (6.3.3), enumerators can directly collect many physical vari

ables (weight, number, size etc.) more accurately. Automatic scales, through whi ch a continuous stream of fish passes, can record the weight of fish mechanicall y or through computerised sensors. Similarly, mechanical or automatic weighing b ins for whole frozen or defrosted fish, prior to entry to a processing line or c old store, can be used to record weights for each batch. Otherwise, boxes need t o be counted and sub-sampled to ensure their fish contents are correctly identif ied and weighed. Fish is often landed in bul together with non-fish materials (e.g. ice, brine s lurry, pac ing material and pallets). It can be very difficult to estimate the t otal fish weight, let alone weight by species,