Chapter-2: Fundamental concept on Research (Research Methodology)-Note-PU

 

CHAPTER: TWO

FUNDAMENTAL CONCEPT ON RESEARCH

2.1 Hypothesis

2.2 Sampling, it’s characteristics, types, benefits and problems

2.3 Field work

2.4 Validity

2.5 Reliability

 

2.1-Hypothesis

Hypothesis is a tentative assumption made in order to draw out and test its logical or empirical consequences.

To clarify the concept of hypothesis, some definitions are considered,

According to,

Black and champion: Hypothesis is a tentative statement of something, the valid of which is usually unknown.

Bailey: Hypothesis is a proposition that is stated in a testable forms and that predicts a particular relationship between two or more variables. In other words if we think that the relationship exist, we first state if as a hypothesis and then test the hypothesis in the field.

A hypothesis is thus a statement about the relationship between two or more variables which needs to be investigated for its truth. It is basically a working assumption.

 

Characteristics/Attributes/Qualities of good Hypothesis

We know that a hypothesis is a proposed explanation for a particular phenomenon. It is usually considered the principal instrument in research. The result of hypothesis are not certain and hundred percent true. A good hypothesis possesses the following certain attributes.

1. A hypothesis should be simple specific and conceptually clear, i.e. there is no place for ambiguity in the construction of hypothesis as ambiguity will make the verification of hypothesis almost impossible.

For example: The average age of male students in the class is higher than that of female.

2. A hypothesis should be capable of verification, i.e. methods and techniques most be available for data collection and analysis. There is no point in formulating a hypothesis if it cannot subjected  to verification. However this does not necessary mean that we should not formulate a hypothesis for which there are no methods of verifications during our research work.

3. A hypothesis should be operationaizable, i.e. it can be expressed in terms of something that can be measured. If it cannot be measured, it cannot be tested and no conclusions can be drawn.

4. Hypothesis should be focused on a specific problem.

5. Hypothesis should state the condition and circumstances under which it is suppose to apply. The context and study units must be cleared.

 

 

Functions of hypothesis:

Hypothesis are inevitable in scientific research. They have the following function to perform.

1. The formulation of hypothesis provides study with focus. It tells us what specific aspects of research problem to investigate.

2. A hypothesis tells us what data to collect and what not to collect.

3. As hypothesis provides a focus, the construction of hypothesis enhances objectivity in the study

4. A hypothesis may enable us to add to the formulation of theory. It enable  us to conclude what is true and what is false.

 

Testing of Hypothesis:

Testing of hypothesis consists of three phases:

a) Construction of hypothesis.

b) Collecting data and evidences.

c) Analysis data and evidences to draw conclusion.

                

 

Sources of hypothesis A hypothesis may be formulated through a number of sources. Following are the main sources of hypothesis. 1.    -Previous study 2.     -Personal experience 3.     -Imagination and thinking 4.     -Observation 5.     -Scientific theory 6.     -General culture   2.2-Sampling Sample: Sample is the collection of units which has been specially selected to represent a large population with certain attributes of interest. Which  population  the sample should be? How large the sample should be. What is mean by specially selected.   Population: Population is the collection of specified group of human being or non-human identities such as objects, educational institution, geographical areas. Population may be finite or infinite. This group usually has some common characteristics that define it. In sampling theory population means the large group from which the sample are taken to draw conclusion. A parameter is a measure that describe the whole population.   Census: The measurement of examination of every element in population is census. Census gives more reliable data when, -The population is very large. -Quick results are required. -In minimizing the cost and time of inquiry. -When the population is hypothetical.   Sampling: Sampling may be define as the selection of part of population on the basis of which judgement or inferences about the universes made.   Types of sampling: Sampling methods There are many sampling techniques that can be used depending on the research question, the population of interest and other factors. 1. Random / probability sampling a.) Simple Random Sampling -Simplest method and forms the basis of all other methods. -The units are selected in such a way that each and every units of population has a equal chance of being selected. -The selection of an item depends on chance and not on judgement of the investigator. Therefore this method is also known as method of chance selection. Merits: -It is free from personal biased of the investigator. -Relatively cheap and simple and can be easily performed in short period of time. Demerits: -Can be used when we have complete list of population, such list always not available. -Becomes costly and time consuming when the field of inquiry is very large and scattered. -Results will be miss leading if the size of sample is not sufficiently large.   b.) Stratified Random Sampling -In simple random sampling, the population to be sampled is treated as homogenous and individual elements are drawn at random from the universe. But if the population is heterogenous with respect to variable or characteristics under consideration, stratified random sampling is generally applied in order to obtain representative sample. -The population is divided into various homogenous groups or strata on the basis of certain characteristics so that various strata are non-overlapping. Then a simple random sampling is used to select a sample from each strata. These samples are then combined to form a single sample of the universe. -The procedure where we have stratification first and then random sampling technique is known as stratified random sampling. Merits: -The investigator first uses his/her judgement to divide the population into various strata. The selection is then done by random sampling technique. So this method enjoys that benefit of both the judgement method and simple random method. -In many field of highly skewed distribution, stratification is very effective and valuable too -As different strata are taken into account, each group of the population will be adequately represented in the sample. Demerits: -Results will not be reliable if each stratum does not contain homogenous unit. -Will not be effective if different strata are overlapping . -Does not help when data are needed about options with a low probability of choice in population.   c.) Systematic Random Sampling -Such method of sampling in which the first sample unit is selected at random and the remaining units are automatically selected at fixed equal interval from one another. -Successfully used when the complete and upto date information of sampling units is available. Merits: -The method is very simple and convenient. -Saves time, money, and efforts. -More efficient than simple random sampling if the list of is complete and unit are serially arranged at random. Demerits: -If the list of units is not arranged in random order result may be misleading. -In some cases the sample will not represented the universe.   d.) Clustered Sampling -Methods of random sampling in which the population is divided into groups called cluster in such a way that characteristics with in the cluster are heterogenous and between the cluster are homogenous, so that the number of sampling units in each cluster should be approximately same and then a simple random sampling technique is applied. These individual cluster are the representative of population as a whole. -The difference stratified random sampling and cluster random sampling is that stratified random sampling is used when each group has same variation within itself but there is wide variation between the groups whereas the cluster sampling is used when there is considerable variation with each group but the groups are essentially similar to each other. Merits: -Elements selected by well designed cluster sampling procedure is easier, faster, cheaper, and more convenient than simple random sampling and stratified random sampling. -Used when the population under study is infinite, where a list of units of population does not exist, when the geographical distribution of units is scattered or when sampling of individual units is not convenient for several administrative regions. Demerits: -Sampling efficiency of cluster sampling methods is likely to decrease with increase in cluster size. -Not to be recommended for taking samples from private residential homes, business and industrial complexes due to widely varying number of persons are households.                                                                             Stratified Random Sampling                    Cluster Sampling   e.) Multistage Sampling -It is the development of cluster sampling. -It is a case of random sampling but sampling is done in various stages. -At first stage universe is divided into large sampling unit and sample is selected at random from them. -At second stage the sample selected at first stage are divided into smaller sampling units again from which a further random sample is taken. -Sampling in any other stages may be done in same way till we get the required result. Merits: -Quite convenient when the area of investigation is large. -As the sample size is reduced at each stage, this method saves time and cost. Demerits: -Generally less accurate other sample of same size which has been selected by a suitable single stage method. 2. Non-Random / Non-Probability Sampling a.) Judgment/purposive/Deliberate Sampling -Samples are taken at the judgement of researcher as that fulfils his/her purpose. -It refers to the sample selected on the basis of what some experts thinks.   b.) Convenience Sampling -In this method selection of sample depends of the convenience of researcher. -The sample then would not necessarily be representative.   c.) Quota Sampling -Derives its name from the practice of assigning proportion of kind of people to interview. -Some Quota are divided to each grow and then samples are selected. -Chances of biasness exist.   d.) Snowball Sampling -Snowball sampling is a non-probability sampling method where new units are selected by other units to form the part of sample. -Also known as chain sampling or network sampling. Snowball sampling begin with one or more study participants, it then continuous on the basis of retrials from those participants. This process continues until the desire sample is achieved. -It is the useful way to conduct research about people with specific characteristics who might otherwise be difficult to identify.  E.g. A person with a rare disease.   Advantages of Sampling -Save time and resources. -In-depth study can be carried out. -Easy to work in limited scope. -Limited number of surveyor are used, there it is easy to provide direction to the surveyor.   Disadvantage of Sampling -Difficult to define sample size and characteristics. -Does not gives good result it sample is biased. -Difficult to judge, the sample and good representation of population or not. -Sometime it is difficult to rely on sample and difficult to convenience people about finding and validity of research.   Sampling Error -Sampling survey involves the study of small portion of population and drawing insure about the population and such there will be on insecurely or errors. Such errors are known as sampling errors. -Those errors which arise due to the use of sample survey due to chance are sampling error. If the censes is taken, sampling error could be expected disappear. The magnitude of sampling error depends upon the nature of the population and the size of same. Sampling error is inversely proportional to the size of sample.   Non-Sampling Error -That portions of difference that can be specifically assign to carelessness in the design of sampling scheme on in planning and of survey, on in collection, processing and analysis of data. -Non-sampling error occurs in all survey, whether it be census or a sample. -Larger the size of sample, larger will be the non-sampling error.   Sampling Biased -Sampling biased is caused by mistake made when defining population. -Sampling biased differ from the sampling. -Sampling biased effect not only the variability under the mean of the estimated parameter but the value themselves, thus has more sever distortion of the surveyor result.   Type 1 Error It is the rejection of null hypothesis when it is true.   Type 2 Error It is the acceptance of null hypothesis when it is false.   2.3-Field Work Organized, systematic, data based scientific investigation in specific situation under taken with the objective of gathering information that enables one to gained familiarity with the situation and generate more knowledge about phenomena under investigation. Key points to be considered during field work 1.) Selection of field -Field area should be identified clearly. -It should be relevant to topic. àIt should be easy geographically, socially and environmentally. 2.) Preparation for field work a.) Preparation of individual surveyor -Surveyor should be mentally and physically prepare. -Should be clear about study and scope. -Should be compatible to work in adverse situation. -Should be expert in his/her field. b.) Preparation of equipment that may include stationary material, maps, camera, clothes, food and medicine c.) Preliminary survey -Preliminary study of concerned literature, inquire with the experts of that field. -Rehearsal of field study. d.) Report -The surveyor should not be biased. -He/she should talk less and listen more. -He/she should not be politically influenced. -He/she shall have a good moral character.   e.) Use of keys information -Key information about the society or field of study that may contribute for reliable data. f.) Note Taking -All information should be noted. -The surveyor group should not be distributed and interrupted during interview.   Activities involved during Field Work 1.) Pre-Field activities -Selection of study area that depends upon interest of the surveyor, capability and fusibility of the study. -Selection of study skill such as case study or fusibility study. -Selection of organization. -Preparation for plane, data collection method and instruments. -Consulting library for more information. -Taking feedback and suggestion from the experts.   2.) Field Work Activities a.) Initial phase -One should be introduced  him/herself to the organization. This includes meeting with the stoke holders and collection of relevant material. b.) Observation phase -This is the practical of field work. One should study and observe the action of the organization. They need to collect relevant data and observe the work system environment. c.) Concluding phase -Rapping of the collected material or observed system.   3.) Post Field Work Activities Final stage of field work means one needs to be prepared for report writing. It includes: -Organizing data in a meaningful way. -Presenting observations wherever appropriate. -Writing the field work report in the recommended format and reporting the field work report. -Submitting the report.   2.4-Validity The concept of appropriateness and accuracy as applied to a research process is called validity. As inquries can be introduced into a study at any stage, the concept validity can be applied to the research process as a whole or to any of its steps. Broadly there are two prospectives on validity: a.) Is the researcher investigation providing answers to the researcher questions for which it was undertaken. b.) If so is it providing these answers using appropriate methods procedures. Validity refers to the truthfulness of findings. It determines whether the research truly measured when it was intended to measure or how truthful the research results are. Validity is also define as the degree to which the researcher has measure what he/she has setout to measure.   2.5-Reliability The extends to which results are consistent over time and then accurate representation of total population under study is referred to as reliability, and if the results of the study can be reproduced under a similar methodology, then the research instrument is consider to be reliable. Relation between Validity and Reliability 1.) High reliability but low validity -The indicators measure something consistently but not the intended concept. 2.) High validity but low reliability -The indicators represent the concept but doesn’t produce consistent measurement. 3.) Low validity but high reliability -It is the worst case where the indicators neither measure the concept nor produce consistent result of whatever they measure. 4.) High validity but low reliability -The indicators consistently measure what they intend to measure.