RESEARCH PROBLEM , HYPOTHESES AND VARIABLES VARIABLES
1. INTRODUCTION The research problem is often argued as the heart of the research process, without which no research process can take place. In formulating the problem the variables must be eminent and easily identifiable while a hypothesis should accompany each research problem. Once a research problem is clearly defined, it should be translated into a research hypothesis that states; a relationship between two or more variables in one (or more) more) populatio population(s) n(s).Thu .Thus s the assignm assignment ent will reflect reflect on the linkage of the research problem, research hypothesis and the existing variables. Firstly, the research problem will be explored, its importance, sources of the research problem, considerations as well as steps to be followed when formulating a research problem. Secondly, Secondly, identification of variables will be discussed using the three common types of variables and the four prominent scales of measurement.Lastly,the definition ,functions,characteristics,types and errors in testing hypothesis will be discussed.
2. THE RESEARCH PROBLEM A research research problem is the first step and the most important requirement requirement in the research process process.. It serves serves as the foundation foundation of a research research study thus if well formulated formulated,, you expect a good study to follow. According the Kerlinger; in order for one to solve a problem, one must know what the problem is. The large part of the problem is knowing what one is trying to do. A research problem and the way you formulate it determines almost every step that follows in the research study. Formulation of the problem is like the input into the study study and the output is the quality quality of the contents contents of the researc research h report. A research problem is often accompanied by research question(s). A Research Research Question is a statement that identifies the phenomenon to be studied. For example, “Wha “Whatt resou esourc rces es are are help helpfu full to new new and and mino minori rity ty drug drug abus abuse e rese resear arch cher ers? s?”” (www.theresearchassistant.com ).
Bless et al (2006:29) write that, in some cases research problems or questions are defined too vaguely and too generally. An important point to keep in mind when defining or formulating a research problem is that it should be specific rather than general. When a problem or question is specific and focused, it becomes amore answerable research quest question ion than than if it remain remained ed gene general ral and unfoc unfocus used ed.. A well well formu formulat lated ed proble problem m is already a half-solved problem .A research problem is expressed as a general question about the relationship between two or more variables. The formulation of a problem introduces the necessity of defining clearly all concepts used and of determining the variables and their relationships. If you know a lot about the topic, you can develop a research question based on your own knowledge. If you feel you don't know much about the topic, think again ( www.esc.edu/esconline www.esc.edu/esconline). ).
2.1THE IMPORTANCE OF FORMULATING FORMULATING THE RESEARCH PROBLEM Accor Accordi ding ng to Kuma Kumarr (2005 (2005:40 :40)) resea researc rch h formu formulat lation ion is like like the the iden identif tifica icatio tion n of a desti destina natio tion n befor before e under undertak taking ing journe journey y. As in the abse absence nce of a desti destinat nation ion,, it is impos impossib sible le to ident identify ify the short shortest est-or -or indee indeed d any any –route –route in the the abse absence nce of a clear clear research problem, a clear economical plan is impossible. A research problem is like the foundation of a building. The type and design of a building is dependant upon the foundation of a building. 2.2 SOURCES OF RESEARCH PROBLEMS Kumar (2005:41) writes that most research in the humanities revolves around four Ps thus People, Problems, Programs, and Phenomena
2.3 CONSIDERATIONS IN SELECTING A RESEARCH PROBLEM According to Kumar (2005:43) there are a number of considerations to keep in mind when when sele select ctin ing g a rese resear arch ch prob proble lem/ m/to topi pic. c. Thes These e cons consid ider erat atio ions ns are; are; inte intere rest st,, magnitude, measurement of concepts, level of expertise, relevance, availability of data
and and ethica ethicall issue issues. s. Baile Bailey y (1982 (1982:19 :19)) is of the conte contenti ntion on that that ther there e are are six six facto factors rs affe affect ctin ing g prob proble lem m sele select ctio ion. n. Thes These e fact factor ors s are; are; the the soci sociol olog ogic ical al para paradi digm gm,, the the researcher’s values, the degree of reactivity inherent in a particular method deemed appropr appropriate iate for gatherin gathering g data, data, the researc researcher’s her’s methodology methodology,, the unit of analysis analysis chosen (small (small or big) and time factor thus, whether the study deals with a cross section of the population at cone point in time or is a longitudinal study conducted over time. Thus Thus it can can be dedu deduced ced that that a hand handfu full of consid consider erati ations ons and factor factors s shoul should d be considered when selecting a research problem. 2.4 STEPS IN THE FORMULATION OF THE RESEARCH PROBLEM According to http://wiki.answers.com, the following are Steps involved in formulating a research problem: 1. Identify a broad area of interest in your academic /professional field. 2. Dissect the broad area into sub-areas by having a brain storming session with your colleagues. 3. Select Select the sub-are sub-area a in which which you would like to conduct conduct your your researc research h through through the process of elimination. 4. Reverse the research questions questions that you would like to answer through your study. study. This can be after formulation of the objectives of can lead you to the formulation of the objective. 5. Assess these objectives to ascertain the feasibility of attaining them in the light of time and other issues like finances and human resource expertise.
3. IDENTIFYNG IDENTIF YNG VARIABLES The following working definitions of variable can be outlined;
An entity that varies from one observation to the next, an empirical property that is capable of taking two or more values(Bless et al 2006;30) al 2006;30)
A property that takes on different values(Kumar 2005:56)
Thus from the above definitions it can be deduced that a variable is a measurable factor that can assume more than one value, hence age,income,gender,pay,job satisfaction can be classified as values. Considering the statement below as an example; The The impa impact ct of remu remune nera rati tion on on the the rate rate of turn turnov over er in Sout South h Afri Africa can n loca locall Municipalities: The case of Buffalo City Municipality . The terms remuneration and turnover represents variables because both can assume two two or more more valu values es whil while e impa impact ct is a conc concep ept. t. Conc Concep epts ts are are ment mental al imag images es or perceptions and therefore their meanings vary markedly from individual to individual, wher whereas eas varia variable bles s are meas measura urable ble,, of course course with with varyin varying g degr degrees ees of accura accuracy cy.. Measurab Measurabilit ility y is the main main differe difference nce between between a concept concept and a variabl variable. e. A concept concept cann cannot ot be meas measur ured ed wher wherea eas s a vari variab able le can can be subj subjec ecte ted d to meas measur urem emen entt by crude/refined or subjective/objective units of measurement (Kumar 2005:56). In some cases the researcher might expect a causal relationship to exist between variables, where a variation in one results in a variation in the other (Bless et al 2006 2006:30 :300). 0).T Taking aking the above above exam example ple,, the the amoun amountt of remun remunera eratio tion n of muni municip cipal al employees may result in a variation of the rate at which the employees leave(turnover) the the muni municip cipali ality ty.Le .Leed edy y and and Ormr Ormrod od (200 (2001:2 1:233) 33) are of the opinio opinion n that that when when we investigate cause-and-effect relationships, we are, of course, looking at the extent to which one variable(the cause) influences another variable(the effect).ln an experimental study, study, the researcher in some ways changes, changes, or manipulates, manipulates, one or more variables in the environment and then measures the effects that such changes have on something else.ln else.ln researc research h in the social social sciences sciences and educati education on ,the “something “something else” being being affected is often some form of human behavior.ln medical research, it might it might be people’s physical healthy or well-being.
3.1 TYPES OF VARIABLE
Bless et al (2006:30) write that, independent and dependant are two most important types of variable. Variables may have particular roles in a certain problem. Thus the common types of variables can now be discussed. 3.1.1 INDEPENDENT VARIABLES Bless et al (2006:30 (2006:30)) define define indepen independent dent variabl variables es (indicate (indicated d by IV) as variable variables s infl influe uenc ncin ing g othe otherr vari variab able les, s, ther thereb eby y dete determ rmin inin ing g the the valu values es of thes these e affe affect cted ed variables.Leedy and Ormrod (2001:233) define an independent variable as a variable that the researc researcher her manipul manipulates ates.. Taking aking the earlier earlier example example,, remuneration is the the manipulated value which can influence the behavior of employees. 3.1.2 DEPENDANT VARIABLES Indicated by DV, are variables whose values are influenced by the value of other values (Bless et al 2006:30). Leedy and Ormrod (2001:233) define dependent value as a variab variable le that that is poten potentia tially lly influe influence nced d by the indep independ enden entt varia variabl ble, e, becau because se it is influenced by, and thus to some extent depends on the independent variable. Also considering the example, turnover is the dependant variable because it depends on the other variable (independent) which is remuneration. 3.1.3 EXTRANEOUS VARIABLES Kumar (2005:60) defines extraneous variable as several other factors operating in a real-life that may affect changes in the dependant variable. These factors, not measured in the the stud study y may may, may may incr increa ease se or decr decrea ease se the the magn magnit itud ude e or stre streng ngth th of the the relation relationship ship betwee between n indepe independen ndentt and dependa dependant nt variabl variables. es. Conside Considering ring the same same example as in above above apart from remuneration remuneration influencing influencing turnover turnover some some factors like working hours, working environment, career growth may also have an influence even though they are not being investigated. 3.2 TYPES OF MEASUREMENT SCALE Measu Measurem remen entt is esse essenti ntial al to an enqui enquiry ry espe especia cially lly when when it is scien scientif tific. ic. Steve Stevens ns (1978:52) in Kumar (2005:67) classified the different types of measurement scale into
four categories namely; nominal or classificatory scale, ordinal or ranking scale. Interval scale and lastly ratio scale. 3.2.1 Nominal Scale A nominal scale enables the classification of individuals, objects or responses based on common/shared property or characteristic for example gender can be classified into two sub-categories; sub-categories; male and female. Similarly citizens in The Republic of South Africa Africa can be classified into two main sub-categories: the First economy and the second economy. 3.2.2 Ordinal Scale The ordinal scale has all the properties of a nominal scale plus one of its own. The simplest ordinal ordinal scale is a ranking. When a market researcher researcher asks one to rank 5 types of beer from most flavourful flavourful to least flavourful, flavourful, he/she is asking one to create an ordinal scale of preference. There is no objective distance between any two points on your subjective subjective scale. For you the top beer may be far superior to the second preferred beer but, to another respondent with the same top and second beer, the distance may be subjectively small. An ordinal scale only lets you interpret gross order and not the relative positional distances. Ordinal data would use non-parametric statistics. These would rank
include:
Median
and
order
mode correlation
non-parametric analysis of variance. (http://www.csse.monash.edu.au ) 3.2.3 Interval The standard survey rating scale is an interval scale. When you are asked to rate your satisfaction with a piece of software on a 7 point scale, from Dissatisfied to Satisfied, you are using an interval scale. It is an interval scale because it is assumed to have equid equidist istant ant point points s betw betwee een n each each of the scale scale elem element ents. s. This This mean means s that that we can interpret differences in the distance along the scale. We contrast this to an ordinal scale where we can only talk about differences in order, not differences in the degree of order. Interval Interval scales are also scales which are defined defined by metrics metrics such as logarith logarithms. ms. In
these cases, the distances are not equal but they are strictly definable based on the metric used. 3.2.4 Ratio Scale A ratio ratio scale scale is the top leve levell of measu measure reme ment nt and and is not not often often availab available le in socia sociall research. The factor which clearly defines a ratio scale is that it has a true zero point. The simplest example of a ratio scale is the measurement of length (disregarding any philosophical points about defining how we can identify zero length). The best way to contrast interval and ratio scales is to look at temperature. The Centigrade scale has a zero point but it is an arbitrary one. The Fahrenheit scale has its equivalent point at -32o. -32o. (Phys (Physici icists sts would would proba probably bly argue argue that that Absol Absolute ute Zero Zero is the zero zero point point for temperature but this is a theoretical concept.) So, even though temperature looks as if it would be a ratio scale it is an interval scale. Currently, we cannot talk about temperature
-
and
this
would
be
needed
if
it
were
a
ration
no
scale.
(http://www.csse.monash.edu.au )
4. HYPOTHESES FORMULATION FORMULATION Hypotheses, Hypotheses, though important, are not essential for a study. study. A perfectly valid study can be condu conducte cted d witho without ut const construc ructin ting g a single single hypoth hypothes esis. is. Ther There e are are howev however er many many defin definiti itions ons of a hypot hypothe hesis sis but but for the purp purpose ose of this this assig assignm nment ent the the follo followi wing ng definitions can be given;
… a conjectural statement of the relationship between two or more variables (Kerlin 1986:17).
… a tentat tentative ive assu assump mptio tion n or preli prelimi mina nary ry statem statemen entt about about the relati relation onsh ship ip between two or more things that needs to be examined(Welman et al 2005:12 al 2005:12
… tentative statement about something, the validity of which is usually unknown (Black and Champion 1976:126).
… a proposition that is stated in a testable form and that predicts a particular relationship between two (or more) variables.ln other words, if we think that a relationship exists, we first state it as a hypothesis and then test the hypothesis in the field (Bailey 1978:35).
Thus from the given definitions it can be deduced that a hypothesis is a premade statement of the results of an investigation indicating the relationship between two or more variables that awaits verification. 4.1 FUNCTIONS OF A HYPOTHESIS Hypotheses are important for bringing clarity, specifity and focus to a research study (Kumar 2005:79).Hypotheses does so by exposing ;the specific aspects of a research problem, important data to a study, focus and finally may enable the researcher to add to the formulation formulation of theory. theory. Hypotheses may only add to the formulation formulation of theory after they have been proven to be true. 4.2 CHARACTERISTICS OF A HYPOTHESIS Kumar Kumar (2005:75 (2005:75)) outlines outlines four major major aspects aspects of a hypothes hypothesis. is. Firstly Firstly a hypothes hypothesis is should should be simple, simple, specifi specific c and conceptu conceptually ally clear clear.Second .Secondly ly,it ,it should should be capable capable of verification.Thirdly,it should be related to the existing body of knowledge and lastly it should be operationalisable thus expressible in terms that can be measured. 4.3 TYPES OF HYPOTHESIS Broadly, there are two categories of hypothesis namely, the research hypothesis and the alternate hypothesis. The research hypothesis is the basis of an investigation and is usual usually ly indica indicated ted by H1.T H1.The he altern alternate ate hypot hypothe hesis sis there there to expli explicit citly ly spec specify ify the the relationship that will be considered true if in case the research hypothesis proves to be wrong. 4.4 ERRORS IN TESTING A HYPOTHESIS
Incorrect conclusions conclusions about the validity validity of a hypothesis hypothesis can be reached if; the study study design selected is faulty, the sampling procedure adopted is faulty, the method of data coll collec ecti tion on is inac inaccu cura rate te,, anal analys ysis is is wron wrong, g, stat statis isti tica call proc proced edur ures es appl applie ied d are are inappropriate inappropriate and conclusions conclusions drawn are incorrect incorrect .There are two common errors when testing a hypothesis. The first error involves rejection of a null hypothesis when it is true Type 1 error error. The and and this this is calle called d Type The secon second d error error involv involves es accep accepta tance nce of a null null
hypothesis when it is false and this is called Type 11 error (Kumar (Kumar 2005:79) .Thus when testing a hypothesis especially when more numerical figures are involved, it is essential essential that every step should be approached with utmost discreetness or else Type 1 or 11 errors may occur.
5. CONCLUSION The research problem is the heart of the research process hence can not be selected in a vacuum thus some considerations and factors have to be considered. Sources of research research problems problems revolve revolve around around the four P’s, P’s, People, People, Problem Problems, s, Program Programs, s, and Phenomena. Formulation of a research problem is a process thus involves following steps. Variables are measurable factors that can assume more than one value and three three types types of importan importantt variabl variables es can disting distinguish uished; ed; indepen independen dent, t, dependa dependant nt and extraneous variables. Nominal scale, ordinal scale, interval scale and ratio scale are the different types of measurement scale which we use to classify variables. A hypothesis is a tentat tentative ive statem statemen entt of the result results s of an invest investiga igatio tion n indic indicati ating ng the the relati relation onsh ship ip betw between een two two or more more varia variable bles s that that awaits awaits verif verifica icatio tion. n. Hypo Hypothe these ses s enab enable le the rese resear arch cher er to add add the the form formul ulat atio ion n of theo theory ry if prov proved ed to be true true.. The The rese resear arch ch hypothesis and the alternate hypothesis are the most important when carrying out or analyzing analyzing research process. However errors my occur when testing hypothesis, hypothesis, Type 1 and Type 2 errors are the common errors. Conclusively a research problem is vital to the research process, and each problem will always variables whose relationship is expressed in the research hypothesis, thus there is a close relationship between the research problem, the variables and the hypothesis.
Reference List Babbie, E .2007. (11 th edition).The edition). The Practice of Social Research .Belmont: Thompson Wadsworth. Bailey, K.D.1982. (2 nd edition).Methods edition).Methods of Social Research, London : The Free Press. Bless,C Bless,C and Higson Higson-Smi -Smith,C th,C.200 .2006.(4 6.(4 th edition)Fu edition)Fund ndam ament entals als of Soci Social al Resea Researc rch h Methods: An African Perspective. Cape Town: Town: Juta&co ltd. Brynard, P.A and Hanekom, S.X.2007. Introduction to Research in Management and Related Academic Disciplines .Pretoria: Van Schaik. Kerlinger , Fred Fred N. (197 (1973) 3) Foundat Foundations ions of Behavio Behavioral ral Researc Research h . 2nd edition. edition. Holt, Holt, Rinehart and Winston.
Kumar, R. (2005).Research (2005). Research Methodology: A Step by Step Guide for Beginners . London, Thousand Oaks, New Delhi: Sage Publications. Leedy, P.D and Ormrod, J.E.2001 (7 th edition).Practica edition). Practicall research: research: Planning and Design. Design . New York: York: Prentice-Hall. Pre ntice-Hall. Welman.Krug Welman.Kruger. er.Mitchell.2 Mitchell.2005 005 (3 rd edition).Researc edition). Research h Methodology Methodology . South Souther ern n Afric Africa: a: Oxford University Press. Terre Blanche, M., Durrheim, K. &Painter, D.1999. (2 nd edition).Research edition).Research in Practice: Applied Methods for the Social Sciences . Cape Town: University of Cape Town (UCT) Press. http://wiki.answers.com/Q/What_are_the_Procedures_used_in_Formulating_a_Researc h_Problem as it appeared on 24 Mar 2009 13:09:13 GMT (accessed on 15 April 2009). http://www.csse.monash.edu.au/~smarkham/resources/scaling.htm as it appeared on 9 Apr 2009 07:29:25 GMT(accessed on 15 April 2009).
