Thesis on "Men Are More Likely Than Women to Desire a Long-Term Relationship"

Quoted Instructions for "Men Are More Likely Than Women to Desire a Long-Term Relationship" Assignment:

I will email the instructions in word as well just incase its easier to read. Thanks

Classified Advertisement Research Project

Topic

The hypothesis would be something like: Men are more likely than women to desire a long term relationship. The dependent variable would be "desire for a long term relationship" and its categories would be (1) personal ads with desire for a long term relationships and (2) personal ads with desire for relationships that are not long term. The independent variable would be "gender" with categories (1) male and (2) female. Notice how a variable cannot have one dimension. It cannot be only "male" because there would be no test of difference there--all the cases would be "male."

What*****s Due:

This order should contain the following:

A five page Letter of Intent (Part I)

A two page narrative about the classified ads (Part II)

One page sampling of classified ads (Part II)

Instructions:

PART I:

You decide to get started right away, beginning by outlining what you will need to include in your Letter of Intent. To reiterate from the announcement above, the Letter of Intent should summarize the proposed research in 5 pages or less in which you: introduce the topic of interest, state a clearly formulated and testable (via classified ads) research hypothesis, and lay out a methodology for testing that hypothesis: identify and define (conceptualize) variables of interest, discuss appropriate measurement (operationalization) of primary variables of interest (including a preliminary codebook), and discuss possible sources of relevant classified ads (though you do NOT worry about selecting specific ads for analysis, or sampling, yet). Eventually, you will complete a pilot study analyzing no less than 100 classified ads. This will necessitate actually gathering a sample of ads (Assignment 7), collecting data through a content analysis of those ads and conducting some elementary an*****s (Assignments 11 and 15). But, the actual doing of this research comes later. For now, you just want to lay out WHAT your question of interest is and HOW you propose to go about testing it. At the minimum, you determine that your Letter of Intent should include the following information:

(A) Introduce the topic of interest. You are asking for money to do this research. It helps to have a research question that is interesting, important and/or addresses some problem that you identify. Contextualize this research, why this topic? Why this particular question? What would answering this research tell us? Who might be interested?

(B) State formal hypothesis/hypotheses: This hypothesis should follow the same format as in assignment 4. One clearly stated hypothesis is sufficient, though you may test two hypotheses if you desire (no more than two hypotheses are allowed for this project). It should clearly state the expected relationship between two VARIABLES of interest in a testable manner. If you don*****t write a hypothesis that states the expected relationship between two variables, it*****s hard to do the rest of the assignment right. For example, in the original research from which this assignment is adopted Simon Davis hypothesizes that, for heterosexuals, men will be relatively more interested in the physical attractiveness of potential partners than women. (Note that the hypothesis makes no mention of classified ads per se, since ads in this case are simply the source of data for testing a more general hypothesis about the relationship between gender and interest in physical attractiveness).

(C) Identify the variables of interest: Clearly state your dependent and independent variables of interest, which are implied in the hypothesis/hypotheses. Each hypothesis would have one independent variable and one dependent variable (ignore any variables other than those implied in your hypothesis). In the above, the variables of interest are gender (independent) and interest in physical attractiveness (dependent). Men, on average, are expected to score higher levels of *****interest in physical attractiveness***** than are women.

(D) Conceptualize each variable of interest: To conceptualize a variable, you define what you mean by that variable in the abstract. Sometimes this is fairly straightforward and deserves only passing mention (e.g. gender). But often, variables can be defined many different ways. So, continuing the example from above, if the variable is *****interest in physical attractiveness,***** what exactly is physical attractiveness? The term physical certainly seems to help limit things here by implying that personality features would not be considered as relevant, but what types of physical features are attractive? Is attractiveness relative or absolute? Any time an ad ***** refers to desired physical traits in a partner, would those be a-priori *****attractive***** or does the ***** need to refer to specific types of physical traits? If specific types of physical traits, which ones qualify and why? If someone wants an *****athletic***** partner and another wants a *****voluptuous***** partner, would they both be coded as attractive or only one or the other? Would an expressed preference for a partner of a particular race constitute physical attractiveness? Certainly, race is physical trait and could thus be considered one element of physical attractiveness.

a. You can certainly do well on this assignment and project even if all of your variables are fairly straightforward conceptually. But, more complex conceptualization of one variable, if done well, will garner additional credit.

(E) Operationalize each variable of interest: Operationalization tells both what to look at and how to categorize what you see into a set of exhaustive and mutually exclusive categories so it will be a variable. This often involves refining what you are interested in so that it is something observable. You choose a level of measurement, decide on the categories, and then carefully spell out what observable cues you will use for categorizing classified advertisements. Try using examples of phrases from actual classified ads, but not the sample you will be coding (basically, you will either count how many times or in how many words, phrases or sentences an ad says something or you will categorize some content. If you categorize, your operationalization should focus on defining the differences among the categories. If you count, your operationalization is focused on defining exactly what content you are counting (e.g. words, phrases, sentences) or otherwise measuring. Ideally, this would be explained in sufficient detail that someone other than yourself could easily follow these instructions and record the variable the same way you would (reliability)

a. Be sure to indicate if your variables are nominal, ratio or ordinal measures.

b. At least one of your variables must be complex enough that it takes at least four sentences to explain the details of how you operationalized it. The other one (or more) variables can be obvious and just briefly explained.

c. When you operationalize the variable you lay out

i. The potential response categories (attributes). So, for the variable *****physical attractiveness***** what are the potential attributes you would like to have? There are at least three options for coding this variable (though certainly you could think of more). One way to code physical attractiveness would be with two (dichotomous) nominal attributes (yes/no*****¦ really three, if you include don*****t know/can*****t tell to be exhaustive) that record whether or not the ad-***** indicates an interest in physical attractiveness in potential partners. Second, you could develop an ordinal scale of physical attractiveness with attributes/response values such as: not at all important, not very important, no preference, somewhat important, very important. Obviously, you*****d need to describe how you code ads according to these categories. Third, you could compute a simple ratio measure by counting the number of relevant references (# of words, phrases or sentences) to physical attractiveness in an ad.

ii. Keywords that might indicate the presence of that variable. The second part of operationalization is to list some keywords or phrases that would be coded as physical attractiveness and, often just as importantly, some that will not (e.g. if your data source requires respondents to list certain preferences you may choose to exclude this information from analysis as ALL ad *****s would have*****¦).

iii. Any further directions for how to go about coding cases, such as particular elements of classified ads that may be considered relevant or not (e.g. you may choose to code, or not to code, mandatory elements of personal advertisements that exist in some listings).

(F) A codebook listing each of the variables included in your study, and only those variables, should be developed as part of the operationalization as well. For each variable the codebook lists the potential response values (or attributes) and the numerical codes that will be assigned to represent each of those values. Finally, it should say a few words about how you come to determine which value to assign a case. In other words, for each variable, specify the criteria (words, phrases, and acronyms) of each value. For example, if your dependent variable is *****physical attractiveness,***** than that term might be operationalized as a nominal variable with three response values (attributes). The codebook indicates what the numerical codes you assign mean as well as how to determine which codes to assign and could look like the following:

Physical attractiveness

01 physical attractiveness in a partner is important. Assign this code if the ad indicates a desire for partners who are *****physically fit,***** *****handsome,***** etc*****¦

02 physical attractiveness in a partner is unimportant. Assign this code if*****¦

03 don*****t know/can*****t tell. Assign this code if*****¦

Professor*****s example:

Variable Code Variable Code

Age: *****number of self-identified years from 18 to 40 with partial years rounded down***** Relationship preference : (operational definition goes here)

18 to 29 1 companionship 1

30 to 40 2 long term 1

relationship 1

Gender fun and games 2

male 1 sexual intimacy 2

female 2 platonic friendship 3

Visit here for more examples of what codebooks could look like:

http://depts.washington.edu/ampol/navResearch/agendasproject.shtml

http://academic.csuohio.edu/kneuendorf/content/hcoding/democb.htm

Say something about potential sampling frames that you have examined. What sources are you considering using to draw your sample of ads? Describe the source. Is there any order to how the ads are listed? Can you say anything about possible readership demographics? Does that imply anything about demographics of classified ad *****s in this source? Compared to classified ad *****s generally? To the general public? For example, the NY Times is a really expensive newspaper to advertise in (high circulation number and national readership) and has readers that are more educated, wealthy, and older than the population at large. Do you think a sample drawn from this source is likely to affect the generalizability or reliability of your results? Maybe compare a couple of possible ad sources in terms of how good you think they will be, what info they have, how generalizable and reliable? Not a lot is expected of you here, we will talk more about the issues of generalizability, reliability and sampling in the coming weeks. For now, the point is to get you looking at some actual ads and thinking about these issues.

PART II:

Rather than an informal discussion about possible sampling frames of interest as in assignment 5 you will (1) identify and describe the actual sampling frame you will use in your study, (2) draw the actual sample of ads for analysis, and (3) describe the methodology used to collect the sample and discuss what it is appropriate. To identify your sample, begin by locating an appropriate sampling frame. Your sampling frame can be any source of classified advertisements you are able to locate. But, whatever source you select needs to have at least 100 entries, so many small local papers may not have a sufficient number of classified advertisements (unless you use multiple editions). How is the sampling frame structured or organized? Are the advertisements organized by sex of seeker and desired partner, alphabetically, by interest, etc? How will the structure of the sampling frame influence the way you sample ads? How representative and generalizable are the results of a sample derived from this sampling frame likely to be? (important note: even if the ads appear *****randomly***** listed, do not assume this to be the case.)

Once you have your sampling frame, you will want to locate the advertisement sample which you will be coding. You will need a minimum of 100 ads in your sample. The point here isn*****t to generate results with high statistical reliability (remember, larger sample size means higher statistical reliability, smaller +/- range) but rather to ensure that you have enough practice coding cases to see where there are places in your codebook, for example, that may need to be changed before a full-scale analysis can be conducted. Be sure to let me know what sampling method you used: simple random, systematic, or stratified with proportionate or disproportionate sampling. Why is this the best sampling strategy, given your hypotheses? How did the sampling frame influence the way you drew your sample? How did your sampling procedure affect the generalizability and reliability of results?

When you have completed this assignment, submit a narrative describing the sampling frame, identifying the particular type of sampling procedure performed, and describing in detail the steps taken. This should be complete enough that I could replicate your procedure. For example, if you selected ads randomly did you just point with a finger, use a random number table, calculator function, or what? Be sure to address the questions above. Also turn in a copy of the sampling frame of advertisements with those ads included in your sample highlighted. Retain a copy of this sample for yourself!

Unfortunately, the details of how you submit a copy of the sampling frame to me is going to depend upon the accessibility of your datasource and your degree of computer literacy. If your classified ads come from a newspaper you may be able to scan into the computer a few pages with ads highlighted and to send along as a pdf file. If you are using an online service such as Yahoo personals, however, your search results might include more than 100 web *****pages***** each displaying 10 individual ads. In such cases where the sampling frame is simply to large to include in its entirety, at minimum I would need an overview of the sampling frame and copies of at least a portion of it, with ads selected for inclusion in the sample highlighted.

I can be flexible on the specific details (though there are better and worse ways to present information on a sampling frame and sample), at this point the key is that (1) it is clear you understand the difference between a sample and sampling frame and that (2) I can confirm that, e.g., rather than select the first 25 ads in each category within a sampling frame (e.g. the first 25 men seeking women ads, 25 men seeking men, etc*****¦) that you actually select 25 randomly, or systematically, or however you claim. I should be able to understand exactly what you did by reading your narrative and be able to confirm by looking at your sampling frame.

Professor*****s additional instructions to part II:

You are asked to (2) take three steps regarding sampling. Follow that section of the assignment (the entire second paragraph) in its entirety. Then, follow the instructions in the third paragraph. The Babbie textbook is a VERY good source of information on sampling. Use it.

Finally, you are asked to (3) submit a narrative about the sampling you did. YES, SUBMIT THIS and be sure to describe IN DETAIL what you did. Refer back to the second and third paragraphs and respond to all the questions that are posed. These are not rhetorical questions, they require answers, and include all of that information in the narrative that you submit. Tell me what you did, how you did it, and why you chose to do it that way.

The description in the instructions of the sampling submission to the drop box is a bit confusing. Your narrative needs to be so clear and complete that I know EXACTLY what your sampling technique is. I want you to submit the narrative and ONE page of classified ads. If you are working with online ads, you likely can cut and paste and resize to fit more than one ad on a page. If you are working with print ads, try scanning one page and send that. But your narrative should be so informative that I need no more than that.

*****