eChapters

In previous sections we discussed about the validity and estimation of validity. Now the question is whether the reliable and valid measuring instrument is sufficient condition to ensure the reliable and valid results in actual experiments? What if the measuring instruments are perfectly reliable and valid but the experiment is conducted in less controlled conditions? Can we expect accurate predictions under such conditions? The answer is No.

Reliability aims at assessing the accuracy and consistency of whatsoever the measuring instrument measures irrespective of whether we are measuring what we are supposed to measure or not. Whereas the validity emphasises on the question “are we measuring what we were supposed to measure?”

To clarify the difference between reliability and validity, let us take an example of instruments that measure colour intensity. Assume, an expert develops two separate instruments X and Y to measure the intensity of green and red colour, respectively. But, the commercial manufacturer wrongly tags the red colour measuring instrument as X. When someone uses this wrongly tagged instrument for measuring the green colour intensity, whatever intensity values the instrument yields are reliable but not valid. It is so because the instrument is measuring red colour intensity accurately (hence data is reliable) but it is not measuring what it was supposed to measure –the green colour intensity (hence data is not valid).

In previous section, we discussed about the concept of the reliability and its importance. In this section, we will discuss about different types of reliability and procedures for estimation of the reliability. We already discussed that one cannot measure the reliability but of course it can be estimated.

Reliability deals with the dependency, consistency, stability, predictability and accuracy of the research data. In our day to day life, when we say that a person is reliable then it means his behaviour is predictable, his behaviour (the way he behaved in past, he is behaving now and the way he will behave in future) will be consistent and he is dependable. On other hand an unreliable person’s behaviour is unpredictably variable and he cannot be depended upon.

There can be no scientific studies which do not involve the process of measuring. Measurement process is the integral part of research activity. Measurement can be defined as the assignment of numerals to the objects or events according to set of rules.  The emphasis in this definition is on phrase ‘set of rules’. If the rules can be set up based on some rational or empirical basis, then theoretically anything can be measured. It is very much easy to understand the concept of measurement in case of physical sciences (Example weight, length, mass etc.) but it is harder understand the concept of measurement in case of psychological objects/phenomenon (intelligence, aggressiveness, attitude etc.) without understanding the predetermined rules of measurement. A rule is a guide, command that tells you what to do/how to do.

Variable is any character or attribute of the given object/set of objects which varies from object to object or from one point of time to another point of time. For example height of individuals is a variable; it varies from person to person.  Kerlinger (1973) defined variable as a symbol to which numerals or values are assigned.

Importance of variables: In previous sections, we discussed about formulating the research problem and development of research hypothesis. Variables are the building blocks of any scientific research problem and research hypothesis. When we consider a hypothesis to be relational hypothesis, we mean that it make claims about relationship between two or more variables. When we test hypothesis, as such we do not test the hypothesis, rather we test the significance of hypothesised relationship between two or more variables. Since variables play central role in observation, measurement and testing of relationship, it is of prime importance to understand the meaning of variables, different types and the way we define these variables.

Most of the scientific research studies –except few descriptive survey research studies–will have one or more hypotheses to be tested as a part of study. Hypothesis is a prediction intended to be tested in a research study (Aron, Coups, and Aron, 2012). Hypothesis is the guide which directs the purpose and process of research by indicating the dependent and independent variables to be observed, extraneous variables to be controlled and the active variables to be manipulated.

Whenever we design a scientific study it is designed in a specific contextual background to solve one or another scientific problem. Subsequently it also involves a set of variables/phenomenon operating in a particular context, set of relationships among these variables and the conditions under which they operate. Certainly, to address an unsolved scientific problem we need to carry out research by employing well defined research methodology and have an idea about possible outcomes.  For example when a extension scientist is asked to study the impact of a novel programme, he needs to understand the background and the context under which the programme was introduced, what are the developmental inputs given under the programme, to whom, when (variables), with what purpose (context), what kind of impact he/she is supposed to investigate (presumed relationship), what methods he is going to use to investigate the impact of developmental programme (research methodology) and what are the possible outcomes of the investigation and how programme interventions are related to the outcome (relationship).

Rationally selected and contextually defined research problem is the foundation to the successful research project. If the selected research problem is not in line with the interests of the researcher, available time, resources, infrastructure & knowledge base then achieving success in research process remains highly uncertain. Consequently, in addition to hindering of professional growth of the researcher, it also results in wastage of time, money and human effort. Most importantly, it may sometimes add up incorrect knowledge into scientific base, hindering the process of scientific progress. Hence the identification of the research problem and designing the problem statement is the most important step in the research process.

Term ‘Scientific research’ is too broad to be explained by using one or two definitions. One should go through the different expert views about scientific research in order to internalize the concept of scientific research. In essence a scientific research is characterized by its rigor, testability, replicability, precision, confidence, objectivity, generalizability and parsimony.

Research: Uma Sekaran (2006) defined research as an organized, systematic, data-based, critical, objective, scientific inquiry or investigation into a specific problem, undertaken with the purpose of finding answers or solutions to it.

Storage of paddy seed lots after harvesting is one of the crucial aspects of processing shopfloor. Paddy due to increased mechanised harvesting may loose vital seed quality parameters. Since seed growers cannot buy a combined harvester themselves-they resort to hiring of this machine. Sometimes in the hurry of completing harvesting they harvest the crop at un-safe moisture content. This un-safe moisture content may go up to 50-60% also. Harvesting of seed crop with high moisture has adverse effect on storability. Ensuring safe seed moisture is very critical especially during rabi-summer seed production cycle. Most of the growers depend on sun drying for reducing the seed moisture. But paddy should not be dried >3% in one cycle. Follow tempering and convection cooling after every cycle and repeat the cycle till seed attains safe moisture content (14%).

The management of storage in open and closed storage is critical before start of any seed processing activities. When seed lots reach the shop floor, they need to be stored properly. Unlike commercial produce seed lots cannot be stored in bulk. These need to be stacked lot wise and there is problem of “stackburn”-resulting from successive layers in building a bag stack. In crops like soyabean-”stackburn” can have detrimental effect on seed quality. The quality parameters like germination and viability can get lost instantly if not properly stacked on the pallets.

To assess the quality of the seed is necessary before it is shipped for packaging and made available to the farming community. To assess certain routine seed quality parameters which are statutory for labelling seed sampling and testing are very critical. Seed testing is not only a mandatory trade requirement but also decides the market potential of a particular product in the market and safeguard the brand image of the company. Seed sampling for labelling and other trade requirements is done at processing plant. The reliability of seed testing is depends on the accuracy with which the sample represents the lot and precision of the laboratory test. In many cases variations in the test results are due to variation in the sampling. Hence, seed sampling is one of the important activities of the shop floor or processing unit.

There are stages in seed supply chain where seed losses happen at harvesting and post harvest stage. These losses may occur due to higher moisture at harvest, due to harvesting methods, processing methods, at seed storage, during transport etc. Estimation of these losses is an important function of seed inventory management. Estimating seed losses help to improve processing and seed distribution efficiency. These simple statistical tools will help to measure quantitative post harvest seed losses. Seed testing can be used for measuring qualitative post harvest seed losses.

The “falling number” is a measurement used by plant breeders and grain quality specialists to determine the soundness of wheat, which is primarily influenced by the degree of sprouting in a grain sample. Hagberg Falling Number is the full name of this international standard of sprout measurement.  The term “falling numbers” refers to the amount of time it takes for the Hagberg steel ball in this sprout analysis tool to fall through a flour/water slurry that has been heated to release the starch from water. 

PPEs are the personal protection equipments which protect the personnel against some hazards in the shop floor. PPEs form important aspects of Environment (safe disposal of treated chemicals, dusts coming out of processing), Safety (to seed storages) and Health (personnel) (ESH) strategy of any shop floor.

NAME OF THE UNIT: 

    1.    Location of Seed Storage 
    2.    Date of Inspections/visit

Seed processing efficiency (SPE) is one of the important aspects in seed processing. It will help in measuring the potential of the shop floor. It will also help in standardizing activities like screens and improving physical purity. SPE will help to improve the quality of the seed material and help in utilizing the machines at their optimum capacity. SPE may vary among the different crops and even among the varieties.
 
To quantify SPE we need to pass the test sample (bulk-pre processed) through the target machine and enter the details in a processing register. To quantify the SPE we need to follow these mathematical formulae:

According to Harrington’s rule (1960) of thumb, for storage, every one percent increase in moisture content, the seed life is halved far seeds of moisture content of 5-14%. For temperature, every 5^ºC rise in temperature, between 0-50 ^ºC, the seed life is halved.

Mechanical damage in oil seeds like soya bean can be induced at any time during harvesting, handling, conditioning and sowing operations. There is relationship between moisture and mechanical damage. In general the low moisture containing seeds are more prone for mechanical damage than the high moisture containing seeds. In this experiment we look for this relationship by measuring seed quality parameters like germination and leakage (electrical conductivity).