Research Methods PDF

Mass Media Research Dr Yasmeen Alyeldeen What do you think is ? this Your attempt to discover what is this is called: “Research” It has many names, but it is most commonly known as the bleeding tooth fungus plant Regardless of how the term “Research” i...

Mass Media Research Dr Yasmeen Alyeldeen What do you think is ? this Your attempt to discover what is this is called: “Research” It has many names, but it is most commonly known as the bleeding tooth fungus plant Regardless of how the term “Research” is used, it means an attempt to.discover something.We all do this everyday Consider the number of things you must analyze, test, or evaluate, to perform :daily tasks Decide which clothes to put on that are.appropriate for the day’s activities Decide when to leave the house to reach your.destination on time Evaluate the best way to tell a friend about a.problem you have  The fact is that when we make any of these decisions, we have to conduct a countless number of tests, or rely on information from previous.tests.In essence, we are all researchers already Children conduct research studies to determine* which items are hot and which are cold, how to ride a bicycle and which persuasive methods.work best with parents Teenagers test ideas about driving, dating, and*.working Adults test ideas about family, finance, and*.survival.What we do everyday is an informal research For the research to be formal -> it has to be.scientific and for the research to be scientific, it has to follow certain steps in carrying it out and must have.five characteristics  The word research was originally taken from a :French word “recherché” which means we search for answers for specific questions and in order to get these answers we need to gather.data The Mass Media Research Phases  The word research was originally taken from a :French word “recherché” which means we search for answers for specific questions and in order to get these answers we need to gather.data  :A Scientific Research Is an organized, objective, controlled, qualitative or quantitative empirical analysis of one or more.variables  :Fields of Scientific research :Research is divided into two major sectors Academic and Private.Which are also called basic and applied  :Academic research Is conducted inside universities, scientific institutions or for the academia in general (done.for an academic purpose)  :Private research.It is conducted by private sectors or businesses.Example: Vodafone (surveying the target audience)  Private Academic 1) Findings are never 1) Usually it is published. published (should be hidden from the competitors.) 2) There is no specific There is a specific )2 budget..budget 3) Time-limit: open (you You have time limitations )3 have no time limitations). (there is a time plan and they are stick to a deadline to finish the procedures.) 4) Basic research as it Applied Research; Do not )4 relies on theories. rely on theories; there is a These two types do not work separately but they are linked to each other. Private research sometimes depends on previous studies carried out in the academic.  Characteristics of scientific research 1-Scientific research is public. For the research to be scientific it should be available for all others who are interested. Therefore, your findings should be available for other researchers to read and use.  2- Scientific research is objective. This means that the researcher should not be biased and follow the objectivity requirements. A scientific researcher deals with facts and realities.not only opinions and points of views  3-Scientific research is systematic and cumulative. Systematic: means that you have to go through.certain steps (follow an order) to carry it out Cumulative: means that we take advantage of the findings of others. We start our work from where.others have stopped  4-Scientific research is empirical (has to be tested and examined). It means that you have to rely on some kind of field work or experience in order to gather your data.and information  5-Scientific research is predictive (rely mainly on predictions). In scientific research we have to use a lot of “Mays” -> (if this happened, then this may.happen)  Different Types of Research Predictive Analytical Descriptive Exploratory The aim of Analytical Descriptive exploratory Predictive research research can be research is research often extends the used to describe undertaken when Is to expect Descriptive and classify few or no previous future approach to The studies exist. The ,possibilities suggest or explain characteristics of aim is to look for based on close why or how ,the subject patterns, analysis of something is e.g. hypotheses or available ,happening number of people ideas that can be information.e.g.watch talk shows tested and will underlying causes form the basis for of an.further research.action  Research process in flow chart Review of literature Review concepts and theories Formulate Define hypothese Research s problem Review previous -3- -1- research findings -2- Interpret Analysis Design and Collection research data (test report of data (including hypotheses -7- -5- ) sample -6- -4- Replicatio n -8-  The typical research process consists :of 8 steps Select a problem (Research Topic))1 Review existing research and theory(when relevant) )2.Develop Hypotheses or research questions)3.Determine an appropriate methodology/ research design )4.Collect relevant data )5.Analyze and interpret the results)6.Present the results in an appropriate from )7.Replicate the Study (When necessary) )8 Selecting a problem (Research )1 Topic)  ?How to select the Research topic :Recognize these 3 concepts when selecting the topic Area Field Topi c Example: Area of study: Mass communication Field: R&TV Topic: Effects of TV violence on children behavior You have to narrow your topic down as much as you can to be able to identify your research problem  From where to select Research ?topic Secondary Sources Primary Sources Any published or unpublished- A document or record containing- work that is one step removed from first-hand information or original the original source, usually data describing, summarizing,.on a topic analyzing, evaluating, derived from, or based on primary source.materials :Example :Example Diaries, letters , photographs Textbooks, review articles, works of art, maps, films biographies, historical films, music sound recordings, Interviews and art, articles about people and Newspapers, magazines,.events from the past advertisements, recollections, Speeches, laws, government records, Everyday situations = Observation  How to select a research topic ? Where to start 1-Academic professional journals :  Periodicals provides you with ideas such as:  Journal of broadcasting & electronic media ,  Journalism and mass communication Quarterly ,  journal of advertising ,  Journal of advertising research  Communication education  Communication research  Journalism quarterly  Public opinion quarterly  Journal of communication  Journal of marketing research  : Academic professional journals-1 They provide you with the latest research topics (12-24months old),and also they suggest topics. that need further investigation 2-Research Abstract : They contain summaries of research articles published in academic journals:  Communication abstract  Sociological abstracts  Psychological abstracts  Dissertation abstracts 3- Internet : IT brings the world to you 4- Mass communication databases like : Ebscohost - proquest 5- every day situation Via radio and TV programs , news papers , movies magazines and personal discussions , these can be good sources of topic. Examples :  Why specific programs are so popular (like oprah programs ).  Who watch weather forecast  Who is the audience of video clips.  So the questions can become research ideas , also you must " pay attention to things around you and conversations with others "  Make sure to ask yourself these 8 :questions after selecting the topic ?Is the topic too broad)1 ?Can the problem really be investigated )2 ?Can the Data be analyzed)3 ?Is the problem significant)4 ?What costs and time are involved in the analysis )5 ?Can the results of the study be generalized )6 ?Is the planned approach appropriate to the project )7 ?Is there any potential harm to the subjects )8  ?Is the topic too broad)1 Beginner researchers frequently choose topics that are too broad.to cover in one study :Ex The effects of television violence on children. (this is a broad topic) ?How to avoid this problem Researcher has to write down the proposed title as a starting -.point.Attempt to analyze the topic into many questions- So, in this example the researcher has to ask the following :questions ?What types of television program ?What the specific kind of violence  Can the problem really be)2 ?investigated whether all the terms of the proposed study can be defined. Remember that all measured variables.must have operational definitions ’A researcher interested in examining youngsters use of the media must develop a working definition.of the word youngsters to avoid confusion  ?Can the Data be analyzed)3 A topic does not lend itself to productive* - research if it requires collecting data that cannot be measured in a reliable and valid style another consideration is the researcher’s previous* experience with the statistical method selected to analyze the data; that is, does the researcher ?really understand the proposed statistical analysis Researchers need to know how the statistics -.work and how to interpret the results All too often researchers design studies that involve advanced statistical procedure they have never used  ?Is the problem significant)4 It is important to determine if a study has merit* before the research is started; that is to determine if the study has or theoretical value practical Will the results add knowledge to information already available in the field? What is the real purpose of the study? This question is important because it helps focus ideas. Is the study intended for a class paper, a thesis, a journal article, or a management decision? Each of these projects requires different amounts of background information  Can the results of the study be)6 ?generalized If a research project is to have practical value* beyond the immediate analysis, it must have external validity; that is, it be possible to generalize the results to other situations What costs and time are involved)5 ?in the analysis A researcher may have an excellent idea, but if costs would be prohibitive the research will be cancelled A cost analysis must be completed early on. It* does not make sense to develop the specific designs and the data-gathering instrument for a project that will be canceled because of lack of funds What costs and time are involved)5 ?in the analysis Time is also an important consideration in research planning. Research studies must be designed so.that they can be completed in the time available Many studies fail because the researchers do not allot enough time for each research step, and in ,many cases the pressure of deadlines creates problems  Is the planned approach)7 ?appropriate to the project a close look at every study is required to plan the best.approach :Ex A researcher might want to measure changes in television.viewing habits The researcher could mail questionnaire to a large sample. The costs of printing and mailing questionnaire are very.high :the researcher could use another tool depending on -.The purpose of the study-.type of the questions - In this example researcher could collect his data by.telephone interviews to reduce costs  Is there any potential harm to)8 ?the subjects Researchers must carefully analyze whether their project may cause physical or psychological harm to.the subjects under evaluation ?Will respondents be frightened in any way Will they be required to answer embarrassing questions ?or perform embarrassing acts is there a chance that exposure to the research ?conditions will have lasting effects Before the start of most public sector research projects involving humans, subjects are given detailed statements explaining the exact procedures involved in the research to ensure that they will not be injured in any way The Research Problem )1 Review Existing Research and )2 theory )Literature Review(.Start from where others stopped - Before conducting your study, you need to- - review the available literature to learn what has been done, how it was done, and what results.were found.It will allow you to save time, effort, and money- It helps you in picking up the research-.methodology and sample The review of literature should provide :you with answers to these questions ?What type of research has been done in the area)1 ?What has been found in previous studies)2 ?What suggestions do other researchers make for further study )3 ?What has not been investigated)4 How can the proposed study add to our knowledge of the )5 ?area ?What research methods were used in previous studies )6  The hypothesis and the research Questions After identifying a general research area and reviewing the existing literature, the researcher must state the problem as a workable hypothesis.or research question  What's a hypothesis A hypothesis is a formal statement regarding the relationship between variables and is tested directly. The predicted relationship between the variables is either true or false.  What's a research Question research question is a formally stated question intended to provide indications about something; it is not limited to investigating relationships between variables  ? Research questions or hypotheses Research questions are appropriate when a researcher is unsure about the nature of the problem under investigation. Research questions and hypotheses Examples Elements of research  What's a concept A concept is a term that expresses an abstract idea formed by generalizing from particulars and summarizing related observations  What's a concept A concept is a term that expresses an abstract idea formed by generalizing from particulars and summarizing related observations  ?Why the concept is important concepts are important for at least two reasons. : First, they simplify the research process by combining particular characteristics ,objects, or people into general categories. Second, concepts simplify communication among those who have a shared understanding of them. Researchers use concepts to organize their observations into meaningful summaries and to transmit this information  Whats a construct A construct is a concept that has three distinct characteristics: First, it is an abstract idea that is usually broken down into dimensions represented by lower-level concepts; a construct is a combination of concepts. Second, because of its abstraction, a construct usually cannot be observed directly. Third, a construct is usually designed for a specific research purpose so that its exact meaning relates only to the context in which it is found.  The variables  Variable is the empirical counterpart of construct or concept.  Variables is a phenomena that can be measured in research.  Variable can has more than one value. :Ex Political interest - as a variable- can take on different :values so a person can be very interested - interested - not interested at all-.  The variables Independent variable is the variable that its effect on the dependent variable is tested in the scientific experiment Dependent variable is the variable which is affected by the independent variable. Therefore, its value depends on the effect of the independent variable..  The variables A research study may involve examining the relationship of more than one independent variable to a single dependent variable. Ex: We can study the effects of watching news programs, political knowledge, and Political interest (as an independent variables) on political participation( as a dependent variable). sometimes we can find multiple dependent variables in a single study, which is called a multivariate analysis. The variables The variables names in some international studies  The sampling One goal of scientific research is to describe the nature of population. The process of examining every member in a population is called a census. But in many situations, researchers cannot examine the entire population because :  it takes long time.  its cost is very high.  The sampling so what is the ideal solution in such situations?  Taking a sample from the population.  Sample is a subset of the population.  If a sample is representative of the population, then the results -from the study using the sample- can be generalized to the population.  If the sample is not representative, the results can't be generalized to the population from which the sample was drawn..  The sampling so what is the ideal solution in such situations?  Taking a sample from the population.  Sample is a subset of the population.  If a sample is representative of the population, then the results -from the study using the sample- can be generalized to the population.  If the sample is not representative, the results can't be generalized to the population from which the sample was drawn..  Sample errors Note: The results must be generalized with some caution, because of the error in all sample selection methods. There are two broad types of error present in all research: 1) sampling error, or error related to selecting a sample from a population; and 2) non sampling error, which is error created by every other aspect of a research study, such as measurement errors, data analysis errors,  Sample errors Note: The results must be generalized with some caution, because of the error in all sample selection methods. There are two broad types of error present in all research: 1) sampling error, or error related to selecting a sample from a population; and 2) non sampling error, which is error created by every other aspect of a research study, such as measurement errors, data analysis errors,  Types of samples There are two types of samples Non- probability sample Probability sample They are not based on any* *They are based on mathematical laws of chance mathematical laws of chance *We can generalize the findings Thus, we can’t generalize the* findings *It is free of our personal judgments nonprobability sampling does* not allow researchers to *Probability sampling allows calculate the amount of researchers to calculate the sampling error amount of sampling error.  deciding to use probability or no probability sample depends on 4 :factors 1- Purpose of the study: A nonprobability sample is appropriate if the research study is not designed to generalize the results to the population but to investigate variable relationships or collect exploratory data to design questionnaire.  deciding to use probability or no probability sample depends on 4 :factors 2-Cost versus value: if the cost of probability sample is too high in relation to the type and quality of information collected or the purpose of the study, then a non probability sample is a good alternative. 3- Time constraints: if the available time is limited , it will be more appropriate to use a nonprobability sample because the probability sample is time consuming.  deciding to use probability or no probability sample depends on 4 :factors 4- Amount of acceptable error: In pilot studies where error control is not a prime concern, nonprobability sample is usually adequate.  Types of non probability samples  1- Available sample (Convenience sample)  An available sample is a collection of easily accessible subjects for study, such as group of students in department of broadcasting.  Available samples can be useful in:  Pretesting questionnaires.  Pilot studies.  But:  Available samples don not represent the population.  Available samples are problematic because they contain unknown quantities of error.  Types of non probability samples  2-Volunteer sample Rosenthal and Rosnow (1969) analyzed several studies and found that volunteer subjects, in comparison with non-volunteers, tend to exhibit: - higher educational level. - higher occupational status. - higher intelligence levels. - greater need of approval. They also seem to be: - more sociable. - more unconventional - younger  Types of non probability samples 2-Volunteer sample These characteristics mean that the use of volunteer subjects may bias the results of a research study. Ex: The polls which are conducted on radio and TV stations, the internet, magazines, and newspapers about some local or national issues- depend on volunteer subjects. But we have to know that the results of these polls are only indications, not scientific” proof”.  Types of non probability samples 3- Purposive sample A purposive sample includes subjects or elements  selected for specific characteristics or qualities and exclude those who don't have the same.characteristics or qualities  A purposive sample is not representative of the general population. Ex: Purposive samples are often used in advertising studies when researchers select subjects who use a particular type of product and ask them to compare it with a new product.  Types of non probability samples 4- Quota sample In the quota sample, subjects are selected according to a known percentage. Ex: If a researcher aim to study how internet users differ from non users in their use of television. He may know that 60% of a particular population use internet so the sample would be composed of 60% internet users and 40% non users.  Types of non probability samples Haphazard sample-5   selecting subjects haphazardly based on appearance or convenience, or because they seem to meet certain requirements.  it is a biased sample.  Haphazard selection involves researcher subjectivity.  some haphazard samples give the illusion of a probability sample; so the researchers must be carful.  Ex:  Interviewing every tenth person who walks in a shopping center is haphazard. Why?  because not everyone in the population has an equal chance of walking in to same shopping center.  Types of non probability samples Snowball sampling  This sampling technique is often used in hidden populations, which are difficult for researchers to access.  To draw snowball sample:  The researcher randomly contacts a few qualified respondents.  Then asks these people for the names of friends, relatives, or acquaintances they know who may also qualify for the research study.  These referrals are then contacted to determine if they qualify for the research.  This sample may be completely biased.  Types of probability samples 1- Simple random sample: In the simple random sample each subject or unit in the population has an equal chance of being selected. If the subject or unit is drawn from the population and removed from subsequent selections, the procedure is known as random sampling without replacement. Random sampling with replacement involves returning the subject or unit to the population so that it has a chance of being.chosen another time  Types of probability samples Ex: a researcher, who wants to analyze 10 prime time television programs out of a population of 100 programs, can draw a random sample by the following steps:  Numbering each show from 00 to 99 and put these numbers in a table  Selecting a starting point from the table randomly.  Selecting the remaining 9 numbers by going up, down, left, or right on the table or even randomly throughout the table.  Types of probability samples Ex: a researcher, who wants to analyze 10 prime time television programs out of a population of 100 programs, can draw a random sample by the following steps:  Numbering each show from 00 to 99 and put these numbers in a table  Selecting a starting point from the table randomly.  Selecting the remaining 9 numbers by going up, down, left, or right on the table or even randomly throughout the table.  Simple random sampling Disadvantages Advantages 1. Detailed knowledge 1. A list of the of the population is population must be not required. compiled. 2. A representative 2. A representative sample may not result group is easily in all cases. obtainable. 3. The procedure can be more expensive than other methods.  systematic random sample -2 How to draw systematic random sample? Ex:  To obtain a sample of 20 from population of 100, randomly select a starting point ( if the number is chosen is11) Generate Sampling interval= 100(the  population) ÷20 (the sample size)so sampling.interval ( 100 ÷ 20= 5)  The sample will include the 20 subjects or items numbered 11,16,21,26, and so on.  Systematic samples are used frequently in mass media research.  Systematic sampling advantages disadvantages 1. Selection is easy. 1. A complete list of the population must 2. Selection can be more be obtained. accurate than in a simple random 2. Periodicity may bias the process. It sample. means that the order of the items in the population list may bias the selection process. 4-Cluster sampling  Cluster sampling is used when a researchers can’t get information about the population as a whole, but they can get information about the clusters (groups).  In cluster sampling, the researchers apply several steps to get their samples.  The most common cluster used in research is a geographical cluster.  Cluster sampling is often more economical or more practical than stratified sampling or simple random sampling 4-Cluster sampling  Cluster sampling is used when a researchers can’t get information about the population as a whole, but they can get information about the clusters (groups).  In cluster sampling, the researchers apply several steps to get their samples.  The most common cluster used in research is a geographical cluster.  Cluster sampling is often more economical or more practical than stratified sampling or simple random sampling stratified sampling -3 It is a probability sampling technique wherein the researcher divides the entire population into different strata (layers) depending on different characteristics of the population( gender, age, education, and socioeconomic status and so on.) Then the researcher randomly selects a sample for each stratum (layer) which is proportional to the actual size of the layer. How to Perform Stratified Random ?Sampling :Ex If you work for a small company of 1,000 people and want to find out which type of TV programs they watch more. Use stratified.random sampling to obtain your sample :Step 1 Divide the population to strata depending on specific characteristics (ex: age)  Step2.Make a table representing your strata The following table shows age groups (strata) :and how many people are in each stratum Total Number of Age People in Strata 20-29 160 30-39 220 40-49 240 50-59 200 60+ 180 Step 3.Decide on your sample size For this example, we’ll assume the sample size.is 50 Step 4 To calculate the proportion of people(sample size) from each group (stratum), use the stratified sampling formula (Sample size of the stratum = size of entire sample / population size * stratum size) Number of People Number of People Age in Strata in Sample 20-29 160 50/1000*160=8 30-39 220 11=220*50/1000 40-49 240 12=240*50/1000 10=200*50/1000 50-59 200 9=180*50/1000 60+ 180 Step 5 Drawing random sample from each strata to.select the survey participants Content analysis Definition of content analysis: Walizer and Wienir define it as any systematic procedure to examine the content of recorded information. Uses of content analysis: The studies of content analysis are usually conducted for one of the following purposes: 1- Describing communication content: EX: -A study described trends in depiction of violence in “slasher” movies. 2- Testing hypotheses of message characteristics: Some analyses attempt to relate certain characteristics of the source to characteristics of the messages. Ex: 1-If the source has characteristic A, then his/her messages will contain elements x and y. EX: Smith and Boyson (2002) discovered that rap music video were more likely to contain violence than any other music genre. 3- Comparing media content to the real world Ex: The national commission of the causes and prevention of violence used content analysis data collected by Gerbner (1969) to compare the television violence with real life violence. 4- Assessing the image of particular groups in society Numbers of content analyses have focused on exploring the media image of certain minority or otherwise notable groups Ex: Roy and Harwood (1997) found that the elderly were underrepresented but were  Steps in content analysis 1- Formulating a research question or hypothesis: Well-defined research questions or hypotheses lead to the development of accurate content categories , which helps to produce more valuable data. Questions or hypotheses content categories produce more valuable data. Ex: If we conduct a content analysis of T.V newscasts, we may formulate many questions such as: 2- Defining the universe : It means specifying the boundaries of the body of content to be considered which requires an appropriate operational definition of the relevant population. Two dimensions are usually used to determine the universe for a content analysis: 1-The topic area. 2- The time period. Ex: If a researcher is interested in analyzing the content of popular songs , he must define: - what is meant by popular songs. - what time period will be considered : the past 6 months? This month only? 3-Selecting a sample: On one hand, if there is a finite amount of content, it maybe possible to conduct a census of this content. On the other hand if there is a vast amount of content available it will be not practical to conduct a census. Thus, a sample must be selected Most content analysis in mass media involves.multistage sampling 1- The first stage is usually to take a sample of content sources. :Ex A researcher interested in specific issues addressed in the T.V programs would first need to select a sample from among large number of programs. He can use the technique of stratified sampling through classify the programs according to program type. 3-Selecting a sample: On one hand, if there is a finite amount of content, it maybe possible to conduct a census of this content. On the other hand if there is a vast amount of content available it will be not practical to conduct a census. Thus, a sample must be selected Most content analysis in mass media involves.multistage sampling 1- The first stage is usually to take a sample of content sources. :Ex A researcher interested in specific issues addressed in the T.V programs would first need to select a sample from among large number of programs. He can use the technique of stratified sampling through classify the programs according to program type. 2- The second stage is to select the date. Ex: in T.V Gerbner and his colleagues (1997) demonstrated that, at least for the purpose of measuring violent behavior, a sample of one week of fall programming and various sample dates drawn throughout the year produced comparable results. Riffe, Lacy, Nagovan, and Burkum (1996) examined sample size for content analysis of broadcast news and found that two days per month chosen at random proved to be the 4- selecting a unit of analysis: Unit of analysis is the smallest element of a content analysis. In written content, the unit of analysis might be a single word or symbol, a theme, or an entire article or story. In television and film analyses, units of analysis can be characters, acts, or entire programs. Definition of the unit of analysis is very important to ensure closer agreements among coders Ex: In a study of Representation of race in TV commercials, the unit of analysis is the character. The definition of this unit is: The first three speaking human characters in each ad. 5- Constructing content categories The category system is used to classify media content. There are two ways to construct content categories: 1-Emergent coding : it means that constructing categories is after examination of the data. Ex: A researcher analyzed the content of major newscasts in a TV channel and, after examining the various news, found that they clustered into three major categories:  News type. (political news, economic news, social news, sports news, other)  Geographical location of the news. (national, regional, international, national/regional, national/ international, regional/ international, other.  News values. (impact, timeliness, prominence, proximity, conflict, bizarreness, currency.  2- A priori coding: it means that constructing categories is before collecting Training coders and doing a pilot-6 :study  Placing a unit of analysis into a content category is called coding.  Individuals who do the coding are called coders.  In researches from 2 to 6 coders are used.  Detailed instruction sheet should also be provided to coders.  Carful training of coders is an integral step in any content analysis to get more reliable analysis.  Next, pilot study is done to check intercoder reliability. 7- Coding the content Standardized sheets (coding instrument )are usually used to ease coding. These sheets allow coders to classify the content(analysis unit) by placing check marks or slashes in predetermined spaces. Researchers who work with television generally video tape the programs and allow coders to stop and start the tape at their own pace while coding data. Coding sheet for studying TV cartoons Character Description code sheet.………………………………Program name..………………………A. Character number …………………………B. Character name C. Role 1-Major 2-Mior Inderterminate D. Sex 1- Male 2-female -3 E. Race 1- White 2-African-American 3-Animal Rboot 5-Native American-4 Indeterminate 7- Other -6 (Specify) F. Age: 1- Child 2-Teenager 3-Adult Mature adult 5- -4 Indeterminate  Role Sex Race Age min majo indetermin femal mal Other Indetermina rboot Anim Whit Indetermin Mature- Adul Child or r ate e e (Specify te Native al Africa e ate adult t Teenager ) America n- n Ameri can 8- Analyzing the data: The descriptive statistics such as percentages, means, modes, and medians, are appropriate for content analysis. If hypotheses tests are planned, then common statistics are acceptable. for example the chi-square test is the most commonly used. Reliability A study is reliable when repeated measurement of the same material results in similar conclusion. Intercoder reliability refers to levels of agreement among independent coders who code the same content using the same coding instrument. If the results fail to achieve reliability, something is amiss with: the coders, the coding instructions, the category definitions, the unit of analysis, or some combination of these. To achieve acceptable levels of Coders should receive examples of units of -1 analysis and definition of categories boundaries with maximum.details Train the coders on using the coding -2 instrument and.the category system 3- Conduct a pilot study. Researchers should select a subsample of the content under consideration and let independent coders classify it. Intercoder reliability can be calculated by several methods. Holsti (1969) reports this formula for Reliability= 2M N1+N2 Where: M= number of coding decisions on which two coders agree. N1 and N2 are the total number of coding decisions by the first and second coder. Thus if two coders judge a subsample of 6 units and agree on 5 of them, the calculation is Coder Coder Ite 2(5) B A m = 0.83 6+6 I I 1 III III 2 II II 3 IV IV 4 II I 5 IV IV 6