8709473
Description
Flashcards by Hanin Lewa, updated more than 1 year ago
|
Created by Hanin Lewa
over 7 years ago
|
|
| Question | Answer |
| What is a case study? is it typically qual or quan? | An analysis of a single case/phenomenon or country, political party, politician, political decision, conflict, in-depth typically qualitative |
| What are the different types of cases? | Debates- study of a single country as a single case descriptive- collecting and examining data and no theory theory testing- tests a theory |
| Debates are..? | a study of a single country/area |
| Advantages of cases? | -in-depth analysis/ description - open for replication |
| disadvantages of cases? | -case selection may not be representative -non-theoretical - often difficult to generalize |
| What is comparative research? | Comparing 2 or a few cases -Small n-cases (n is number of cases) -typically qualitative -countries (Federalism in CA, US and Germany) -provinces in a country (Pol party preferences in Alberta, ON, etc) |
| What is the most similar systems design? | Similar systems/countries/cases but significant differences/variation in the issue/outcome being analyzed |
| What are examples of most similar systems design? | Similar economies and political democratic systems in CA and US. Higher union membership rates in CA than the US. No strong Social democratic party in US in comparison to CA. |
| What is the Most Different Systems design? | Dissimilar systems/cases/countries produce similar outcomes. No significant differences in the issue/outcome being analyzed in these dissimar countries/cases/systems. There is a common factor that produces the same outcome. |
| What is an example of most Different Systems design? | Revolutions in France, Russia and China. Theda Skocpol. Healthcare systems in CA and Cuba |
| Criticisms of most Different Systems design are? | Small number of cases, selection of cases, debates in political science, measurement problems/importance of context, existence of only one case/explanatory factor. |
| What is survey research? | Major method of quantitative research -academic and non-academic (elections, government, business, opinion polls) - participation is voluntary -confidentiality of respondent (can't be identified) |
| Standardized/structured interviews- and questionaires? | -Used in survey research -produce standardization in the asking of questions and recording of answers same questions are asked in the same order (translated if needed) |
| Standardized/structured interviews are benefecial because? | -reduce error due to interviewer variability -characteristics of the interviewer may influence responses given |
| What are Questionnaires? | -essentially structured interviews without an interviewer -involves filling out a form and being returned to researcher often through mail, email or online -list of survey Q's and A's -helpful in analyzing survey data |
| What are basic points of conducting interviews? | Question should be asked as stated, answers should be recorded as exactly as possible, Q's should be asked in same order given on interview schedule. |
| why should Questions be asked exactly as stated? | small changes in wording make a big difference |
| What are the 3 types of interview questions? | close-ended- list of answers given, respondent chooses one. open-ended leading question- can affect answers |
| The YES PRIME MINISTER video is an example of what type of interview question? | leading questions |
| What are the types of surveys? | face-to-face, telephone, online, mail. |
| What are face to face surveys? | at home or on the street, etc. Time-consuming, costly, more representative. |
| What are telephone surveys? | Can't exceed 30 mins, cheaper, quicker to administer, but many people in many countries don't have phones |
| What are mail surveys? | Sent by mail, expensive, low response rate, no interviewers needed |
| What are online surveys? | Quick to construct, cheap (survey monkey), interview doesn't have to be transcribed, high dropout rate, not everyone has internet |
| What are omnibus surveys? | surveys comprised of many small surveys. cheap. researcher can ask one or small number of questions into a larger survey |
| What is a census? | -surveys whole population, very expensive |
| What is a population vs a sample? | population- all cases about which one seeks knowledge. EX. adult population of CA, UO students sample- people selected for survey, used in survey research. EX. sample of UO students, sample of adult canadians. |
| What are representative samples? | A sample containing essential characteristics of the population |
| What is a probability sample? | a sample selected using a random process such that each element of the population has a known probability of being selected. But response rates in different groups of samples can differ. need to be weighted to ensure representation. |
| What are the 4 types of non-probability sampling? | Purposive, quota, snowball, and volunteer sampling. -hard to generalize the population, and hard to be representative |
| What is purposive non-probability sampling? | -Non-random participants. example. university students |
| what is a volunteer sample? | self-selected participants |
| what is a quota sample? | set quotas for specific groups |
| What is a sampling error? | errors of estimation that occur as a result of the differences between the characteristics of a sample and those of a population |
| what is a response rate? | percentage of the sample that participates in the study |
| Sample size? | it is the absolute size of the sample that matters, not the proportion of the population that it comprises. As sample size increases, sample errors tend to decrease. - Sample sizes can be smaller in other surveys or sample types |
| As sample size increases... | sample errors tend to decrease |
| What are the typical sample sizes in national surveys? | 1000, 1600, 2500 The greater the heterogeneity of the population on the characteristics of interest, the larger the sample size should be. |
| What are the types of respondent problems in survey research? | Acquiescence, laziness or boredom, and social desirability |
| Explain the 3 types of respondent problems in survey research. | Acquiescence- respondent trying to please the researcher laziness or boredom- preventing respondent from giving a genuine response social desirability- being reluctant on providing a sincere answer if it will make them appear unlikable or bad. |
| What is the relevence of the invention of the computer? | -made quantitative research easier/possible. -gave researchers a resource to input and calculate numbers such as the mode, median, and mean. |
| Who is Atanasoff? (Eyesteelfilm) why is he important? | inventor of the first computer, which helped speed up solutions of problems in mathematical physics and statistics. |
| What are variables? | provide a measurement of concepts and contain different values. |
| What is an example of a variable-value relationship? | political party prefernces in Canada variable have the following values Conservatives Liberals NDP other |
| What are the different types of variables? | Nominal, ordinal and interval-ratio. -They have different levels of measurement/types of variables -specific level is determined by values |
| What is a nominal variable? | a level of measurement describing a variable that has values that cannot be ranked in contrast to other types of variables EX. Gender: Female, male Pol Party: NDP, LIBERAL ETC |
| What is an ordinal variable? | values we can rank-order but cannot find the average value (mean) EX. education: high school, university, post graduate. religiousity: very high, moderately high, moderatley low, very low |
| What are interval/ratio variables? | variable whose values are rank ordered and have equal distances between adjacent values. ex. age (years) income ($) |
| What is a frequency distribution? | percentage distribution of the variable of interest. tables or charts that summarize the distribution of a variable. frequency tables |
| What are the measures of central tendancy? | mean- average (interval-ratio variables) median- middle value (ordinal and interval-ratio variables) mode- value that occurs most frequently (all types of variables) |
| What is a research hypothesis? | a statement between a relationship between the independent variable and the dependent variable that we want to prove or disprove. ex- political parties are affected by level of education |
| What is a null hypothesis? | a statement of no difference between independent and dependent variable. ex. political parties are not affected by level of education |
| what are the dependent variables and independent variables? | dependent variables- a variable that depends and is affected by the independent variable. - we are interested in explaining these independent variables- affect the dependent variable. we use this to explain the dependent variable |
| What is a cross-tabulation table? | also called contingency table - a common way to test hypotheses -used for bivariate analyses. most appropriate for nominal and ordinal variables -dep var in rows indep var in colomns |
| what are tests of statistical difference? | formal and exact way to test a hypothesis. - derived with help is advanced mathematics. relies on inferential stats easier to get statistically significant results in a large sample. |
| What are types of hypotheses testing? | Chi Square test- for nominal and ordinal valuables. used in cross tabulation tables T-test- for interval/ratio dependant variable and nominal independant vraiable with 2 categories correlation analysis- interval/ratio variables regression analysis-interval/ratio variables |
| What is statistical significance? | -if a relationship between a dep and indep variable is significant and in expected direction/pattern. -null hypothesis is false -research hypothesis is supported |
| null hypothesis is true means? | the relationship between dep and indep variable is statistically non-significant -research hypothesis rejected |
| Measures of association? | provide info about the strength of the relationship between variables and direction of association (ordinal or interval-ratio variables) -used in cross tabulation analysis |
| Cramers V? | Chi square based measure of association - best for nominal variables b/w 0-1 (no to perfect association) |
| Gamma? | for ordinal variables -shows direction of association b/w -1 to 1 (perfect neg association and perfect positive association) |
| What is content analysis? | analysis of patterns, themes, biases and meaning of the content - examination and coding of documents and texts ex. newspapers, magazines, films/tv shows, images, social media, etc mixed- can be quan and qual |
| What is the simpsons video and example of? | content analysis |
| What things can be counted or analyzed? | words, subjects and themes (manifest and latent content), positive or negative view |
| What is coding? | designing a coding schedule- the form where the data are recorded. designing a coding manual- instructions to coders, list, categories, numbers, general guidance |
| what are the issues in devising coding schemes? | -categories must be mutually exclusive, exhaustive, instructions to be clear, and pilot test will reduce risks or error. |
| What is the significance of the massacre video example? | importance of latent content |
| Qual content analysis | inductive, interpretative, Research questions, conceptuatization, coding scheme (unit of analysis), small # of cases, coding data, identifying patters, biases etc, checking coding reliability. |
| Quan content analysis? | deductive approach, research hypothesis and questions, conceptualization and operationalization (variables and unit of analysis), coding scheme, data (sample random or non random of population), coding data -human or comp, coding reliability, statistical analysis |
| advantages of content analysis? | quan form - transparent and easily replicated. allows for longitidinal analysis, unobotrsive method, flexible, can be used with several kinds of unstructured info, overcomes social barriers to research acces. |
| disadvantages of content analysis? | problems with content analysed (authenticty, credibility, representativeness), coding reliability issues (intercoder-2 vs intracoder-same), difficult to answer why questions, potential for invalid conjecture, especially in discussions of latent meanings. |
Want to create your own Flashcards for free with GoConqr? Learn more.