1. What’s the problem with how PR people typically consider statistical analysis?
Answer: In the PR world, people in the public relations profession are not very familiar with the idea of statistical analysis because a lot of statistical analysis ins thrown from the marketing aspect. People in the public relations profession either do not know how to do statistical analysis, have not thought of using it, or completely avoid it as an idea to turn to in the world of public relations. In order to turn the statistical analysis from a problem into a pleasure, the article gives some helpful tips, which include the use of data to achieve goals, understanding the data’s weaknesses and strength, taking control of your and making it work, an getting over the fear of data.
2. What are “statistics”? What are they for?
Answer: According to writing at Colorado State, statistics are a set of tools used to organize and analyze data. For example, statistics is used to transform a survey full of non-numbers into a visual that allows for the easy reading of data. Several pieces of data can be compiled into a visual, easy to read image that displays data. For example, in our survey projects, we will be using the survey results as a set of data and be putting it into a layout that we can present to the class that is helpful and understandable for others to comprehend in a smaller time period. Statistics are used for two different purposes, to provide description and to provide prediction for certain variables.
3. What does “prediction” mean?
Answer: Prediction is the idea that from existing data, a general or probabilistic idea can be made about the future. In the sub-idea of generalizability, the data that is used and has gone through the sorting of statistics can be applied to similar concepts. For example, reasons why students are not involved at athletic events or involved in clubs or organizations could result in data that shows reasons of homework; this could apply to other questions asked in similar surveys on campus. Prediction can also be probabilistic; this means that the data that is shown can show a pattern that will continue into the future, such as the U.S. Census and predicting the population projection in 2020.
4. Sampling is a strategy connected to descriptive or inferential statistics?
Answer: Sampling is a strategy connected to inferential statistics. When talking about inferential statistics, a person is refereeing to the idea of the data descriptions giving a pattern that can possibly predict the future if the pattern continues. Sampling takes a portion of the entire population, uses the portion to collect data. After the data is collected, it can be used as a probabilistic idea for the rest of the population. Sampling is connected with inferential statistics because were “inferring” a similar pattern will happen in the future, or with a larger group of people.
5. What are the three kinds of variables: explain.
Answer: There are three different types of variables that exist: nominal, ordinal, and interval variables. Nominal variables are variables that put data into different types of categories. Nominal variables name each of the categories and keep track of the different frequencies of occurrence. Ordinal variables are variables that rank different pieces of data not by numbers, but by terms of degree. There is no numerical difference present between the different types of data points in ordinal variables. Interval variables indicate the numerical value of data, as well as the numerical distance between each data point in a collection of data.
6. Under “methods,” what are the three basic categories of methods? Explain.
Answer: The three basic categories of methods are analyzing individual variables, analyzing relationships among variables, and analyzing differences between groups. Analyzing individual variables consist of looking at the central tendency of each variable and the different measures of variation in a particular variable. Analyzing differences between groups is a second category that looks at the differences of “scores” between two or three groups at a time. The third category is analyzing relationships among variables, which takes a look at the correlation or regression among several variables in a collection of data.
7. What’s the difference between inferential and descriptive statistics?
Answer: The difference between inferential and descriptive statistics is how the data is presented and what questions it answers about the collection of data. Descriptive statistics presents the data in a way that describes what or what is not being shown in the data. Inferential statistics include the presentation of what or what is not in the data; the inferential statistics also goes beyond in trying to find a conclusion and answer questions that existed before the collection of data. Inferential statistics cover more area than descriptive statistics, especially when accomplishing set goals.
8. What is univariate analysis?
Answer: Univariate analysis is the idea of examining one specific variable at a time with three different kinds of criteria. These criteria include the distribution, the central tendency, and the dispersion of a variable. The distribution refers to, “ the summary of the frequency of individual values or ranges of values for a variable.” In other words, the distribution looks the amount of times of certain variables or a difference of value in a variable. Central tendency refers to “an estimate of the ‘center’ of a distribution of values, which include the mean, median, and mode.” The third is dispersion, which refers to the spread of the values around the central tendency. All three of these criteria are used in univariate analysis.
9. What are mean, median, and distribution standard deviation?
Answer: Mean is the most widely used term used to describe the central tendency of univariate analysis. The mean is considered the average, which is computed by adding all values and dividing by the number of value points in the collected data. The median is the data point found in the middle of the set of values when the values are listed from least to greatest. The distribution standard deviation is the considered the “more accurate and detailed estimate of dispersion” because a value that is extreme can change slightly the mean, median, or other method of finding an average of a certain piece of data.